URBAN LAND-USE PLANNING USING GEOGRAPHICAL …
83
LUMA-GIS Thesis nr 35 Kazi Masel Ullah 2014 Department of Physical Geography and Ecosystem Science Centre for Geographical Information Systems Lund University Sölvegatan 12 S-223 62 Lund Sweden URBAN LAND-USE PLANNING USING GEOGRAPHICAL INFORMATION SYSTEM AND ANALYTICAL HIERARCHY PROCESS: CASE STUDY DHAKA CITY
Transcript of URBAN LAND-USE PLANNING USING GEOGRAPHICAL …
LUMA-GIS Thesis nr 35
Kazi Masel Ullah
2014
Department of
Physical Geography and Ecosystem Science
Centre for Geographical Information Systems
Lund University
Sölvegatan 12
S-223 62 Lund
Sweden
URBAN LAND-USE PLANNING USING
GEOGRAPHICAL INFORMATION SYSTEM AND
ANALYTICAL HIERARCHY PROCESS:
CASE STUDY DHAKA CITY
Kazi Masel Ullah (2014). Urban land-use planning using Geographical Information System and Analytical Hierarchy Process: Case study Dhaka city. Master degree thesis, 30 credits in Master in Geographical Information Sciences (GIS) Department of Physical Geography and Ecosystems Science, Lund University
i
Urban land-use planning using Geographical Information System and Analytical Hierarchy
Process: Case study Dhaka city
By: Kazi Masel Ullah Master thesis, 30 credits, in Geographical Information
Sciences
Research Supervisor: Dr. Ali Mansourian Associate Professor
Department of Physical Geography and Ecosystem Science Lund University, Sweden
ii
Acknowledgement
I am now fine. This is not a complex decision making process! It’s very simple. I express my
deep gratitude to my supervisor, Dr. Ali Mansourian, Associate Professor, Department of
Physical Geography and Ecosystem Science, Lund University, for frequent guidance,
overwhelming encouragements and endless patience that made me possible to complete the
thesis work. I wish Dr. Ali would supervise me up to the end of my career.
I am grateful to the teachers of LUMA-GIS (Lund University Master’s Program in
Geographical Information Systems), who put lot of efforts for building the course materials
and for delivering those in intuitive way. The course works through the program built my
deep understanding and problem solving skills to complete a Master’s thesis using
Geographical Information System. I am thankful to the administration and other staffs of the
department as well for their instructions and feedback for any difficult situations. I appreciate
the services I got from Lund University digital library. I am also grateful to Lund University
as well as the Swedish Government for providing such graduate level education in free of
cost.
I would like to thank the experts, whose opinions were shared for my study. I regret I could
not mention their names for ethical reasons. I express my thanks to RAJUK’s authority for
their data supports.
I am thanking fellow students of LUMA-GIS program for their kind cooperation, especially,
the South African lady Monika Lehman Ogden and the Iranian man Aliteza Mollasalehi.
Finally, I express my best gratitude, thanks and heartiest love to my brother, mother, father
and other family members who supported me religiously from my early life to complete my
all level of education.
iii
Abstract
Urban land-use planning, which is a useful tool for the sustainable development of a city, is a
complex decision making process. However, the modern GIS technologies facilitate such
complex jobs in two ways – (i) GIS allows to work with large numbr of datasets, (ii) a
number of methods, techniques or models could be embeded in GIS for land-use suitability
analysis. Analytical Hierarchy Process (AHP), which is a kind of Multi-Criteria Decision
Making (MCDM) technique, could be used for urban land-use planning with support of GIS
technology. The aim of this thesis is preparing urban land-use planning using GIS and AHP,
where the case study is Dhaka city.
Dhaka, which is one of the fastest growing mega cities in the world and is the capital city of
Bangladesh, is facing acute pressure of increasing population and unplanned urbanization,
despite, a number of planning interventions have been taken for the planned development of
the city. Recent Detail Area Planning (DAP) for Dhaka city was a cumbersome job but
brought little benefits. DAP primarily prepared a land-use plan at city scale using GIS
technology. Although huge resources and times were used to build the GIS database, it had
promlems on (i) specifying data requirements, (ii) ensuring quality database (having
topological rules, elimination of sliver polygons etc.) and (iii) using the database for spatial
analysis in view to make better planning decision. In this connection, this thesis tried to
conceptualize a model to build geographical database for urban land-use planning to address
first two problems and applied GIS-based AHP technique for more sophisticated analysis
(problem-iii).
After literature review and selection of the study area (Group-E of DAP), the study set a
number of criteria through sharing experts’ opinions. Based on those criteria the collected
GIS data was transformed into the Geodatabase, where the geodatabase was conceptualized
using Unified Modelling Language (UML). Five experts’ opinions were shared and further
literature reviews were done for calculating Eigen Values using AHP methodological
operations. The Eigen Values show the degree of priority of the criteria. Using Eigen Values,
raster criterion maps were prepared from data available in the geodatabase. These criterion
maps were overlaid to develop a composite map which was later classified to prepare
suitability map.
The research result shows that highly suitable area (13%) should be used for urban residential
zone; moderately suitable area (35%) should be designated as mixed use zone; low suitable
iv
area (42%) should be reserved for agricultural use and open spaces; and not suitable area
(10%) should be protected from any types of activities except agriculture. The research
approached an urban land-use planning at a regional scale. The research results were also
validated with Detail Area Plan of Dhaka Metropolitan Development Plan package in some
order. Such validation concludes that Geographical Information System based Analytical
Hierarchy Process can be applied successfully for preparing urban land-use planning at the
regional level.
v
Table of Content
Acknowledgement…………………………..……………………………………………..ii
Abstract……………………………………..……………………………………………..iii
Table of Content…………………………..……………………..………….……………..v
List of Tables…………………………..……………………..……………..……………..vi
Table of Figures…………………………..……………………..………….………….....vii
Chapter 1: Introduction
1.1 Introduction ................................................................................................................. 1 1.2 Background ................................................................................................................. 1
1.3 Objectives .................................................................................................................... 3
1.4 Methodology ............................................................................................................... 3
1.5 Thesis Structure ........................................................................................................... 4
Chapter 2: Study Area
2.1 Selection of the Study Area .............................................................................................. 5
2.2 Data Collection ................................................................................................................. 5
2.3 Location ............................................................................................................................ 5
2.4 Area and Population ......................................................................................................... 6
2.5 Administrative .................................................................................................................. 6
2.6 Land-use ........................................................................................................................... 7
2.7 Physical and Geological Characteristics .......................................................................... 8
2.8 Utilities and Infrastructure ............................................................................................... 8
2.9 Community Facilities ....................................................................................................... 9
Chapter 3: Conceptualizing Geodatabase
3.1 Database and DBMS ...................................................................................................... 10
3.2 Geodatabase ................................................................................................................... 10
3.3 Conceptual Modeling of Database ................................................................................. 11
3.4 What is the UML ............................................................................................................ 12
3.5 Conceptual Modeling of Geographic Database ............................................................. 16
Chapter 4: MCDM and GIS for Urban Land-use Planning
4.1. Land-use Planning ......................................................................................................... 18
4.2 Land-use Classification .................................................................................................. 20
vi
4.3 Criteria for Urban Land-use Planning ............................................................................ 22
4.4 Multi-Criteria Decision Making Model (MCDM) and GIS ........................................... 23
4.5 The Analytical Hierarchy Process (AHP) ...................................................................... 25
4.6 Why AHP has been used for this study .......................................................................... 25
Chapter 5: Designing Geodatabase and Evaluation of Land Suitability using AHP
5.1 Modeling Geodatabase using UML ............................................................................... 27
5.1.1. Administrative and Planning Boundary ............................................................. 28
5.1.2 Land-use .............................................................................................................. 31
5.1.3 Physical and Geological ...................................................................................... 32
5.1.4 Utilities ................................................................................................................ 32
5.1.5 Community Facilities .......................................................................................... 33
5.2 Data Processing .............................................................................................................. 34
5.3 Selection of Criteria ....................................................................................................... 34
5.4 Preparing Comparison Matrix ........................................................................................ 35
5.5 Estimating Weight for Factors and Criteria ................................................................... 37
5.6 Evaluation of Land Suitability in GIS ............................................................................ 41
Chapter 6: Resulst, Discussion and Conclusion
6.1 Result and Discussions ................................................................................................... 44
6.1.1. The Land-use Suitability Indexes ...................................................................... 44
6.1.2 Land-use Proposal ............................................................................................... 56
6.1.3 Comparison with DAP ........................................................................................ 57
6.2 Conclusion ...................................................................................................................... 58
References…………………………..…..…………………………………..……………..59
Appendix 1…………………………..…………………………………………...………..63
Appendix 2…………………………..…………………………………………...………..69
Published Thesis in LUMA-GIS..…….………………………………………...………..69
List of Tables
Table 1: Sources of data.....……………………………………………………….……………5
Table 2: Administrative units of the study area..……………...…………………………...…...6
Table 3: Detail land-use classes of the study area………………….…….……………………7
Table 4: Gross land-use classes of the study area……………………………………………....8
vii
Table 5: Bartholomew land-use classification…………………………………..…………….20
Table 6: Factors and criteria of land-use…………………………………….……………......35
Table 7: Fundamental scale used in Pair-wise Comparison………..………………………....36
Table 8: Weight values of factors, criteria and factor ratings………………..………………..42
Table 9: Area of different suitability types...……….…………………….…………………...44
Table 10: Land-use classes under each suitability types……………………….……………..44
Table 11: Elevation classes under each suitability types…………………..………………….45
Table 12: Earthquake risk area under each suitability types………………….……………....47
Table 13: Road classes under each suitability types………………………..………………....48
Table 14: Electricity classes under each suitability types…………………..………………....48
Table 15: Gas line classes under each suitability types…………………………..…………...49
Table 16: Proximity to school under each suitability types…………………………..……….51
Table 17: Proximity to healthcare services under each suitability types…………..………….51
Table 18: Proximity to bus stop under each suitability types………………..………………..53
Table 19: Proximity to market/bazar under each suitability types…………..………………..54
Table 20: Proximity to park under each suitability types………………..…………………....55
Table 21: Dominant features under each suitability types…………………………..………...56
List of Figures Figure 1: Flowchart of methodology……………………………………………...…………..4
Figure 2: Study area map……………………………………………………………………...6
Figure 3: Class Structure……………………………………………………………………..13
Figure 4: Association between two classes…………………………………………………..14
Figure 5: Aggregation between two classes...………………………………………………..15
Figure 6: Composition between two classes…...…………………………………………….15
Figure 7: Multiplicity of a class with other…………………………………………….…….15
Figure 8: Generalization between classes……………………………………………………16
Figure9: An example of theme diagram……….……………………………………....…….16
Figure 10: Stereotypes for spatial representations…………………………………………...17
Figure 11: Final class diagram for land distribution theme…………….…………………….17
Figure 12: Broad themes of land-use planning………………………………………………28
Figure 13: Class Diagram - Administrative and Planning Boundary theme….……………...30
Figure 14: Class Diagram – Land-use theme….……………………………………………..31
viii
Figure 15: Class Diagram – Physical and Geological theme….……………………………..32
Figure 16: Class Diagram – Utilities theme…….……………………………………………33
Figure 17: Class Diagram – Community Facilities theme………………….………………..33
Figure 18: Diagram to choose preference level between land-use and Physical/Geological………………………...………………….……………………………..36
Figure 19: Land-use Suitability Index map….………………………….……………………43
Figure 20: Comparison between suitability types and land-use criteria………….………….45
Figure 21: Comparison between suitability types and elevation criteria…………………….46
Figure 22: Comparison between suitability types and fault line criteria…………………….47
Figure 23: Comparison between suitability types and road criteria……….…………………49
Figure 24: Comparison between suitability types and electricity criteria….………………...50
Figure 25: Comparison between suitability types and gas line criteria………….…………...50
Figure 26: Comparison between suitability types and school criteria………….…………….52
Figure 27: Comparison between suitability types and healthcare criteria…….……………...52
Figure 28: Comparison between suitability types and bus stop criteria……….……………..53
Figure 29: Comparison between suitability types and market/bazar criteria………………...54
Figure 30: Comparison between suitability types and park criteria……………….…………55
1
Chapter 1: Introduction
1.1 INTRODUCTION
Urban land-use planning is a useful tool for the sustainable development of a city (Van Lier
1998). Land-use planning tries to formulate activities to be proposed, administer potential
changes and protect incompatible changes. Such administrative and management strategies
through land-use planning guide to ensure sustainability of a city. So, land-use planning is an
essential component of urban planning.
Although it’s a complex decision making process to prepare a perfect land-use planning,
however, the invention of modern GIS technologies has eased such complex jobs in two ways
– (i) GIS allows to work with large number of datasets, (ii) a number of methods, techniques
or models could be embeded with GIS for land-use suitability analysis. A wide number of
social, economic, physical and environmental indicators are considered for better urban land-
use planning. Gepgraphical data in a GIS environment supports to use those indicators in
more sophisticated way in the decision making process of urban land-use planning. However,
for dealing with the datasets in a GIS environment, a geographical database management
system is required, especially, when the datasets are robust and complex. To build such
geographical database, at first, it is essential to prepare a conceptual model so that the data
requirements and their interrelations are well defined, and database can be used to store,
modify and query the data with security. Then, a number of Multi-Criteria Decision Making
(MCDM) models or techniques embeded with GIS could be used for land-use suitability
analysis, where the importance of each indicators of land-use planning are determined in
more sophisticated way through subjective and or objective judgments. From a literature
survey it was found that Analytical Hierarchy Process (AHP), which is a kind of MCDM
technique, could be used for urban land-use planning with support of GIS technology. Thus,
the aim of this thesis is preparing urban land-use planning using GIS and AHP, where the
case study is Dhaka city.
1.2 Background
Dhaka, the capital city of Bangladesh, is one of the fastest growing megacities in the world.
Its current population is around 15 million, which would be around 20 million in 2020 as
projected by World Bank (2007) that would make Dhaka the world’s third largest city. But,
the city is expanding in an unplanned way (Islam et al. 2009). Although a number of plans
2
and control mechanisms have been prepared since 1959, but, the city never experienced
planned development rather rapid haphazard expansion from the city centers to its outskirts.
To control the unplanned spatial development, the city land development and management
authority, Rajdhani Unnyon Kartipakkha (RAJUK), formerly named as Capital Improvement
Trust, adopted Dhaka Metropolitan Development Plan (DMDP) in early nineties of the last
century. It was a paradigm shift in the history of urban planning in Bangladesh. It avoided the
traditional master plan concept which was quite rigid in nature for planning process and plan
implementation (Nallathiga 2009). Besides, it was the first time when Bangladesh tried to
prepare a GIS-based urban plan through DMDP. The development plan was in three tiers –
Structure Plan, Urban Area Plan and Detail Area Plan (DAP). The first two tiers were
completed during 1992-1995. It was supposed to start the work of DAP preparation
immediate after the completion of first two tiers of the DMDP plan package. But due to
budgetary constraints and bureaucratic complexities, the actual works of DAP preparation
were commenced in 2004. The allotted time for the DAP preparation was 2 years, but, the
draft final report of the DAP was completed in 2008 and after going through the review
processes it’s official gazette was published in 2010 (GOB 2010). That is, the DAP got its
legitimate identity after 15 years of the structure plan. However, the DAP has not become
operational in reality where RAJUK claim that they have limited staffs to supervise the plan.
Moreover, the plan period of DMDP will be finished in 2015, where it was the commitment
to finish the DAP within 2000. So, the entire DMDP package seems useless although huge
resources were used in vision of this plan. The DAP primarily prepared land-use plan
according to the structure plan guidelines. Here, GIS technology was used and entire 590 sq.
km of DMDP area were surveyed (physical, topographical and socio-economic survey) to
build the database. To do these tedious jobs huge resources and time were used. So, this
research tried to look an alternative way so that such resource and time taking jobs could be
minimized for such urban land-use planning.
Around 50 datasets on different topographical, physical, land-uses, administrative and other
features were prepared to build the GIS database according to the Term of Reference (ToR)
of the DAP project. The problems in this GIS database were in three folds – i) Actual datasets
required for a comprehensive plan, the types of data, their relationship classes etc. were not
well defined before preparing the database. As a result, during data acquisition, database
preparation and at the moment of plan preparation surplus data, data shortage or data
redundancy prevailed. ii) During preparing database, neither any format of geographical
3
database management system nor any topology rules were followed. So, data is not clean.
Overlapping, sliver polygons and associated problems exist significantly. iii) During taking
planning decisions, databases were usually used for getting map supports only. GIS is unique
and very useful tool to make spatial analysis and modeling using its spatial and non-spatial
data for making better decision. However, in DAP such implication was not followed. So, a
real benefit of using GIS was neglected. In this connection, this study tried to conceptualize a
model to build geographical database for urban land-use planning to address problem (i) and
problem (ii); and tried to apply a GIS-based AHP technique to make more sophisticated
analysis for urban land-use planning so that an example can be shown to address problem
(iii).
1.3 Objectives
Considering the above problem statements, the aim of this research is preparing an urban
land-use plan for Dhaka city using GIS and AHP. To fulfill this aim following set of
objectives were satisfied -
1. Requirement analysis for fixing datasets and for selecting criteria affecting urban
land-use planning.
2. Designing and construction of a geographical database.
3. Prioritizing the criteria using AHP
4. MCDM analysis for land-use planning.
1.4 Methodology
A stepwise methodology was followed in this research. The workflow of the research can be
shown by figure -1. During the preliminary studies a number of literatures were reviewed and
the study area was selected. Based on preliminary studies and considering experts’ opinions,
the requirement analysis was done for setting data requirements and for getting criteria
affecting the land-use planning. Then, the model of geodatabase was designed using Unified
Modelling Language (UML). During designing the model, the criteria of land-use planning
were considered. Then data was collected and was exported into geodatabase under the
designed structure of UML modeling. Five experts’ opinions were shared and further
literature reviews were done for calculating Eigen Values using AHP methodological
operations. The Eigen Values show the degree of priority of the criteria. Using Eigen Values
raster criterion maps were prepared from data available in geodatabase. These criterion maps
were overlaid to develop a composite map which later was classified to prepare suitability
4
map. Finally, using the suitability map land-use plan was suggested. A detail discussion on
how the methodological works have been done will be found in chapter-5.
1.5 Thesis Structure
The thesis is organized in six chapters and two appendices. Chapter 1 is the introductory
chapter, which shows the background, objectives and outline of methodologies of the thesis.
Then, the chapter 2 briefs the justification of selecting study area, data sources and the
demographic, administrative, geographic and other related profiles of the study area. Chapter
3 and chapter 4 discuss the findings from the literature reviews. Chapter 3 highlights the
theoretical discussions on database, geographical database and their conceptual modelling in
Unified Modelling Language (UML). Chapter 4 clarifies the importance of the use of GIS-
based MCDM techniques for land-use planning. In the same chapter it is also explained why
Analytical Hierarchy Model, which is a kind of MCDM technique, was used for this study.
The major methodological operations for the study, which are preparing geodatabase and
application of GIS-based AHP for evaluating land-use suitability, are discussed in chapter 5.
Based on those land suitability indexes, the chapter 6 analyzes the existences of different
land-use criteria on different suitability indexes, makes land-use proposal on the basis of
those results and analytical discussions and finally, compares the proposals with existing
DAP. Then, the thesis makes the conclusion at the end of the same chapter. Besides, experts’
given scores for different factors and criteria are shown in appendix 1 and their combined
factors and criteria weights are mentioned in appendix 2.
Figure 1: Flowchart of methodology
Requirement Analysis
Preliminary Study Literature Review
Study Area Selection
Geodatabase modeling using UML
Sharing expert opinion
Data Collection
Creating Criterion Map
Preparing Composite Map
Preparing Suitability Map Land-use Planning
Selection of Criteria for Land-use Planning
Calculating Eigen Value
Creating Geodatabase
5
Chapter 2: Study Area
This chapter covers some basic information about the study area. The information has been
collected mainly from the survey data and report of Detail Area Plan (DAP) prepared for
Dhaka Metropolitan Development Plan (DMDP) area (RAJUK 2010).
2.1 Selection of the Study Area
There was no single database for DMDP area. The DMDP area was divided into 5 groups
(Group A, B, C, D and E) during preparing DAP. Each group has separate database. It would
be a huge job to combine the five databases into one and to conduct study for whole DMDP
area. So, the area of group E was selected for the study as the area has huge potentiality for
future urban growth.
2.2 Data Collection
Different secondary data were collected for this study. The datasets and their sources are as
the following Table -1 –
Table 1: Sources of data
Datasets Source
Land-use RAJUK
Structure RAJUK
Sport height/contour/DEM Water Development Board
Road networks RAJUK, LGED
Gas supply Titas Gas Transmission and Distribution Company Ltd.
Electric supply RAJUK
Community facilities RAJUK
2.3 Location
The study area is located in the north-western part of DMDP area (Figure-2). The planning
authority of DMDP area, RAJUK, has 26 Strategic Planning Zones (SPZ). For preparing
Detail Area Plan (DAP), the RAJUK’s jurisdiction area was divided into five separate groups
and a number of locations. The study area falls within the Group-E.
6
Figure 2: Study area map
2.4 Area and Population
The study area covers around 75,789 acres. According to Bangladesh Bureau of Statistics
(BBS), in 1991, the population of the study area was 4,16,682, which rose to about 6,42,320
in 2001. During preparing the Detail Area Plan, the projected population of the area for 2015
was estimated as 11,77,272 that would be almost 24% increase over that of 2010.
2.5 Administrative
The study area consists of entire Savar Upazila (sub-district) of Dhaka district and part of
Gazipur district. It has one municipality, namely, Savar Pourashava and has 14 unions (local
rural government unit). Following Table-2 shows the administrative units of the study area –
Table 2: Administrative units of the study area
Administrative unit Name
District Dhaka, Gazipur (partially)
Upazila (sub-district) Savar, Gazipur (partly)
Pourashava (municipality) Savar
Union (local rural Simulia (Part), Tetuljhora, Yearpur, Pathalia, Kaundia,
7
government unit) Dhamsona, Bhakurta, Banagram, Ashulia, Biralia, Savar
Cantonment, Basan (Part), Kayaltia (Part), and Kasimpur (Part)
Police Station Savar
Source: BBS 2001 and DAP.
There are some special or restricted areas within the study area, like as Jahangirnagar
University, Savar Cantonment. These areas are served by their own governing ordinances.
2.6 Land-use
A more detail land-use classification can be found from DAP’s database (Table-3).
According to DAP information, most dominant land-use is agriculture, which comprises
around 40.7% of total area followed by vacant land (20.8%), residential (20.2%) and water
bodies (7.6%).
Table 3: Detail land-use classes of the study area
Land-use Type Proportion of the Area
Agriculture 40.727%
Circulation Network 1.617%
Commercial Activity 0.217%
Community Service 0.003%
Education & Research 1.444%
Forest Area 2.601%
Governmental Services 0.078%
Homestead 0.848%
Manufacturing and Processing Activity 2.499%
Mixed Use 0.092%
Open Space 0.233%
Recreational Facilities 0.006%
Residential 20.239%
Restricted Area 0.886%
Service Activity 0.058%
Transport & Communication 0.071%
Vacant Land 20.787%
Water Body 7.594%
Total 100.00%
8
The land-use data was processed from each parcel of land under cadastral plots. Such huge
detail data could not be used for this thesis work. Because it was not possible to convert that
detail data in raster format from shape file format. Besides, this study purposes to make a
land-use plan at regional level, but not at detail level. So, in this study the land-use classes
were divided into four classes (gross land-use class) based on existing dominant use and
importance of proposed development (Figure-2). The land-use classes applied in this study
and their proportions have been shown in Table – 4.
Table 4: Gross land-use classes of the study area
Land-use Area (sq. m) Proportion of the Area
Agriculture 34531.09 46%
Homestead 600.15 1%
Residential 14970.82 20%
Vacant Land 25730.05 34%
Total 75832.12 100%
2.7 Physical and Geological Characteristics
The soil characteristics of the study area are mainly alluvial soil originated from the
Pleistocene period. The southern part of the study area is composed of alluvial soil of the
Bangshi and Dhalashwari rivers. Main rivers are Bangshi, Turag, Buriganga and Karnatali. A
substantial part of the study area is geographically a part of the Pleistocene Terrace, popularly
known as Madhupur Tract. Such lands are characterized by high, undulated land surface with
red soil, crisscrossed by flood plains and streams. Except the southern part, almost entire land
of the planning area falls in this category.
The elevation of the area ranges from 2m to 13m. Most of the places the slopes are gentle
except some patches of high land. These high lands are free from flood.
Some big tectonic faults exist in Bangladesh. Dhaka city and its surrounding areas are also
located within a seismic zone, so, the risk of earthquake exists in Dhaka metropolitan area as
well as in the study area. Some geologists in Bangladesh claim that the rivers within and
around the city are fault lines.
2.8 Utilities and Infrastructure
There was no systematic and planned development in the study area. The road networks of
the area were not built following proper planning standard. The study area is served by about
9
3190 km of road in which 42 km of national highway, 13 km of regional highway and 3135
km of local and other roads. There is no municipal water supply system in the area. Most of
the cases people use tube-well. All over the study area has electricity connection. Gas supply
is only provided to the urban areas and to the areas through the major highway but the rural
areas.
2.9 Community Facilities
In Group-E area there are one public, one private university, two private medical colleges
along with a number of schools and madrasha (religious schools). There are 33 daily kitchen
markets, 10 hats and 4 wholesale markets within the planning area. The bus stops exist
mainly through the major highways. Bus stops are available after 2-5 km distances. Around
40 healthcare centers could be identified from the DAP survey data. According to the DAP
report, most of the community facilities are below the standard. There are few or no
recreational parks or open spaces in the study area.
10
Chapter -3: Conceptualizing
Geodatabase
This chapter presents relevant concepts on Geodatabase design. As Geodatabase is itself a
database, so, the topic starts from defining database and database management system before
defining geodatabase. Then, it highlights a major requirement for any database building, that
is, the conceptual modeling of database. Later, it explores the usefulness of UML for
designing database. Then, the use of UML in the geographical database framework is
discussed.
3.1 Database and DBMS
A database is collection of data organized in a structured way, so that; information can be
retrieved quickly and reliably (Closa et al. 2010; Yeung and Hall 2007). The invention of
Information Technology has led the database to be used in a management system, which is
called Database Management System (DBMS). DBMS is a set of programs, in other words
software systems that enables following actions to be performed in a database –
• Stores and extracts information from a database
• Modification of database by adding, editing, deleting and sorting of data
• Supports queries and produces reports based on those queries
• Provides facilities to maintain integrity, quality, performance and concurrency within
database
• Ensuring security of database that controls data accessibility and allows protection
and recovery from hardware failures.
There are several models of DBMS. Among them Relational Database Management System
(RDBMS) is most widely used. RDBMS stores data in the form of related table, which is
much efficient for the users to query data in many different ways.
3.2 Geodatabase
Modern GIS uses Spatial Database to integrate the geometry or features data with other types
of data (Yeung and Hall 2007). Spatial database facilitates storing and querying data that is
related to objects in space, including points, lines and polygons. Other typical databases can
understand various numeric and character types of data, while, spatial databases need
additional supports to process spatial data in the form of geometry or feature. Spatial data,
11
which is also called geographical referenced data or geospatial data, focuses cartographic or
mapping perspectives. Some common features of a perfect spatial database system are storing
and managing capacity of both spatial and non-spatial data; maintaining spatial references;
allowing all possible geometrical representations, like, point, curve, surfaces etc.; ensuring
topology for managing spatial relationships among geometric objects and maintaining data
security, integrity and allowing spatial data query. In this sense, ESRI Geodatabase can be
viewed as a good example of a spatial database system.
3.3 Conceptual modeling of database
Real world is much complex. So, we use model as “model is a simplification of reality”
(Booch, Rumbaugh and Jacobson 2005). The model synthesizes important aspects of a
purposeful activity to be modeled and eliminates the unnecessary things (Rumbaugh,
Jacobson and Booch 1999). Model is expressed in any convenient form, for example, model
of a building can be presented by drawing on a paper or 3-D figures made of cardboard.
Database modeling in the software system has similar consideration – abstraction of the
essential elements of the observed reality from nonessential elements (Lisbao Filho and
Iochpe 2008). A conceptual database modeling describes possible data content, structures and
constraints applicable to them. Like other models, to express the database modeling
descriptions in a convenient way, conceptual data modeling language is used. A conceptual
data modeling language is use of formal expressions (notational and semantics) of tools and
techniques used for data modeling.
Use of a conceptual data modeling language is must to prepare the data schema at the
conceptual phase (Lisbao Filho and Iochpe 2008). The schema is the collection of linguistic
and graphical representations that describe the data structure of a database and database
processing operations (Yeung and Hall 2007). A lot of modeling languages have been
evolved (Booch, Rumbaugh and Jacobson 2005). Different modeling techniques used for
database management systems can be classified in the following categories (Yeung and Hall
2007) –
• Hierarchical Systems
• Network Systems
• Relational Systems
• Object-oriented Systems
12
The hierarchical systems and network systems are now out-of-date. Those systems are much
rigid to implement the database. Relational database became much popular for its greater
flexibility. The basic foundation of a relational database management system is ‘tables’,
where each table contains a unique name that distinguishes it from other tables. One or more
column of a table serves as the key (primary and or secondary key) of a table, which relates
one record with another. Traditional databases were only limited to text-based applications.
Growing demand of information other than text data, like, graphics, video, sounds, maps and
elements of other multimedia environment led the development of object-oriented database
systems. The systems mainly evolved from object-oriented programming languages (Yeung
and Hall 2007).
Contemporary views of modeling processes in the field of computer science are based on
object-oriented perspective. The main building block of this approach is the object or class.
Classes are treated as a set of objects. The system of defining objects and classes consists a
large number of occurrences. The relations between these occurrences and their properties or
attributes can be set in this approach (Jacobson, Ericson and Jacobson 1994). Such approach
is easily expandable. A number of authors in late 80s to early 90s tried to formulate different
object-oriented methodologies. But the concepts and ideas of those methodologies were
almost similar. So, later efforts were made to unify all the object-oriented approaches into a
common standardize language (Rumbaugh, Jacobson and Booch 1999). Such endeavor
developed Unified Modeling Language (UML), which got strong tendency for to be adopted
in computer science as well as in database design (OMG n.d.).
3.4 What is the UML
“UML is a graphical language for visualizing, specifying, constructing, and documenting the
artifacts of a software-intensive system” (Rumbaugh, Jacobson and Booch 1999). Following
are UML capabilities those are much useful for human and machine –
• UML is used to understand, design, browse, configure, maintain, and control
information which is essential to construct a system.
• UML includes semantic concepts, notation, and guidelines that offer a standard way
to write a system’s blueprints.
There is no distinct demarcation between various concepts and constructs in UML, but, for
the conveniences they can be categorized into several views. At the top level, views can be
divided into three areas: structural classification, dynamic behavior, and model management
13
(Booch, Rumbaugh and Jacobson 2005). Structural classification describes the things in the
system and their relationships to other things. Dynamic behavior describes the behavior of a
system over time, and Model management describes the organization of the models
themselves into hierarchical units. There are many UML diagrams to model a system under
the above views or similar other aspects. Some UML defined graphical diagrams are –
• class diagram
• use case diagram
• behaviour diagrams:
o statechart diagram
o activity diagram
• interaction diagrams:
o sequence diagram
o collaboration diagram
• implementation diagrams:
o component diagram
o deployment diagram
UML class diagram provides the semantic constructs for a conceptual modeling language of
geographic database (Lisbao Filho and Iochpe 2008). So, here we will only describe about the
class diagram. Discussion about class diagram was made from (Booch, Rumbaugh and
Jacobson 2005; Shekhar and Chawla 2003).
Class Diagram
Class diagrams are the most common diagram found in modeling object-oriented systems. Class
diagrams describe the classes in the system, and the static relationships between classes.
Class:
A class is represented by a rectangle. Figure-3 shows a typical class in a class diagram.
Class Name
Attributes
Methods
Figure 3: Class Structure
14
A class has three rows -
• The first row is the class name. A class is representation of an object or variable that
can be anything like a person, place, thing, concept, event, screen, or a report
applicable to the given system. So, class name is the object name which allows
identifying the object. The class name typically has the first alphabet capitalized. If
there are two or more words in a class name, then first alphabet of each word is
capitalized and is joined with all the words.
• In the second row list of attributes of the class is presented. The syntax is:
- Attribute: Type = ‘default value’
• A list of methods goes in the third row. The syntax is:
- MethodName(List of parameters): Return Type.
Like object-oriented concepts, access modifier can be applied in the class. Access modifier
determines the scope of visibility of the class and its methods and attribute. Documentation
information, like notes and constraints, can be also added to a class.
Relationships between classes:
A relationship is a semantic connection between classes. The attributes, operations and
relationships of a class can be known with another class using semantic notations (symbols)
representing UML relationships. Followings are some common relationships used in UML:
Association:
Association is semantic connections between two or more classes. In an association
relationship, the public attributes and operations of one class are known by another class.
Associations are drawn on a class diagram with a single line. For example, an elementary
school exists within a residential neighborhood boundary. Here it is an association
relationship and can be shown by the following diagram –
Figure 4: Association between two classes.
Aggregation:
Aggregation is a specified association. It is a relation between a whole and its parts.
Aggregation is applied when one class is collection of another class. But the collected class
does not have strong life cycle dependency on the collector class. It is also called “has a”
relationship. Figure-5 is an example of aggregation relationship.
Elementary School
Residential
Neighborhood
15
Figure 5: Aggregation between two classes.
Composition:
Composition is more specified association. It connotes that there is a strong life cycle
dependency between classes. Unlike aggregation, when a part must belong to a whole, then
composition is used. Following diagram is an example of composition.
Figure 6: Composition between two classes.
Multiplicity:
Multiplicity notations are placed near the ends of an association. Such symbols indicate the
number of instances of class linked to one instance of other class. For example, one
residential neighborhood has one or more elementary school, but each elementary school
exists within a particular residential neighborhood. See Figure-7.
Figure 7: Multiplicity of a class with other.
Following Table shows different indicators of multiplicity to express different meanings.
Indicator Meaning 0..1 Zero or one
1 One only
0..* Zero or more
1..* One or more
n Only n (where n>1
0..n Zero to n (where n>1)
1..n One to n (where n>1)
Generalization:
Generalization shows an inheritance relationship. When one class (subclass) is inherited from
other class (super class), then, the super class is considered as a Generalization of subclass.
Generalization allows one class to inherit all of the attributes, operations, and relationships of
another class. In UML, the subclass is treated as child class and the super class is treated as
parent class. An example of generalization for a city database may be Districts. Several types
of districts, like residential districts, business/commercial districts, industrial districts etc.,
may be generalized by the super class District. See Figure-8.
Car Model Engine
Car Model
Elementary School
Residential
Neighborhood
1..*
Car Engine
Car
16
Figure 8: Generalization between classes.
3.5 Conceptual Modeling of Geographic Database
Geographic Database (GeoDB) integrates spatial data and data relationships into database
management system. Spatial Database Management System supports to build, storage,
structure and assist basic operations for spatial data manipulation, while, GIS provides
mechanisms for analysis and visualization of geographic data (Shekhar and Chawla 2003).
Like any traditional database, GeoDB must include conceptual, logical and physical design
phases during its designing stage. Among different modeling languages used for conceptual
modeling of GeoDB, the UML-GeoFrame modeling language is one of the most recent uses
developed by Lisbao Filho and Iochpe (1999). The modeling process based on the UML-
GeoFrame comprises five steps (Lisbao Filho and Iochpe 2008):
• Step 1: to identify themes and subthemes for each target region of the application. See
Figure-9.
• Step 2: to draw a class diagram for each theme specified in step 1, associating classes of
different themes, if this is the case. At this stage, modeling of the data is carried out. For each
theme, the several elements of the real world that is being modeled are abstracted.
• Step 3: to model the spatial characteristic of each geographic phenomenon. Stereotypes are
defined for different spatial representations. See Figure-10.
• Step 4: to model spatial relationships
• Step 5: to model temporal aspects. This step is not widely used.
Figure-11 shows a final class diagram on land distribution theme for a hypothetical system in
view to making agricultural reform.
Figure 9: An example of theme diagram (Lisbao Filho and Iochpe 2008).
District
Residential District
Commercial
District
Environment
Hydrography Climate
Relief Vegetation Soil
Land distribution
Warehouses
Roads Environment Laws
Cropped areas
Municipal district: Geographic Region
17
Figure 10: Stereotypes for spatial representations (Lisbao Filho and Iochpe 2008).
Figure 11: Final class diagram for land distribution theme (Lisbao Filho and Iochpe 2008).
18
Chapter 4: MCDM and GIS for Urban
Land-use Planning
4.1. Land-use Planning
Generally, land-use is determined by the human activities on land. Most clarified definition
of land-use was first pronounced by Burley (1961). Burley defined land-use by two
interrelated phenomena – land cover and land utilization. Land cover describes both the
natural and human altered land surface where human actions take place. While, human
actions determine the land utilization, both phenomena are significantly important for land-
use planning. Because, on the one hand, human actions are limited by the land cover settings
and on the other hand, human activities alter the land cover (Wang and Hofe 2007). The main
purpose of land-use planning is to designate land-use types for an area where change is
expected. Land-use planning help to answer following questions –
1. What should be built?
2. Where should it be built?
3. When and how should it be built?
4. What impact will it generate?
According to Wang and Hofe (2007), answering such questions help to produce ‘good’
community, although, they also found controversy to define a good community. However, the
implication of their questions for large city scale or regional scale may be complicated. It
would make a complex discussion to show how different human activities can be distributed
over the land mass of a city or region.
Human activities can be divided into three major categories—residential (where people live);
employment (where people work); and others (non-residential and non-employment
activities) (Wang and Hofe 2007). Human activities and land have an interrelated function.
Land provides space, materials, energy etc. to meet human demand for performing activities.
On the other hand, human activities are dependent on the availability of land, its
characteristics and resource availability.
From the modern era human activities are growing in manifold and with great intensity. Such
activities are contributing rapid urbanization, which has adverse environmental impacts if the
urban development is not managed sustainably. Land-use planning can play a great role for
19
sustainable development (Van Lier 1998). A basic proponent for land-use planning is to
classify land based on human activities to be proposed and human activities that exist. Some
discussions on land-use classification have been made in the next section. Before that we
need more clarifications on land-use planning.
FAO (1996) defines land-use planning as “the systematic assessment of land and water
potential, alternatives for land-use and economic and social conditions in order to select and
adopt the best land-use options.” So, Land-use planning involves decision making process so
that best alternative option can be attained. Growing urbanization demands expansion of
urban areas, at the same time, alteration of existing land-uses within the urban area. To make
such changes in sustainable manner land-use planning is a useful tool to administer the
potential changes and to protect incompatible changes. Decision in land-use plan formulates
goals and objective for resource management (United States Department of the Interior
Bureau of Land Management 2005). Within these goals and objectives, measurements for
future land management actions are adopted. Land-use planning also assists to device site-
specific implementation mechanism.
According to FAO (1996) guidelines there are 10 steps in land-use planning –
1) Establish goals and terms of reference
2) Organize the work
3) Analyze the problems
4) Identify opportunities for change
5) Evaluate land suitability
6) Appraise the alternatives: environmental, economic and social analysis
7) Choose the best option
8) Prepare the land-use plan
9) Implement the plan
10) Monitor and revise the plan
A crucial stage in land-use planning is the suitability analysis, which is central part of land-
use evaluation. The purpose of land suitability is identifying an area which is suitable for a
given type of land-use. For example, finding the areas which are suitable for residential use.
Nowadays, people are widely using the modern GIS technologies for land suitability analysis.
Scientists, researchers and experts are trying to develop tools and techniques for land-use
evaluation that fit with the social, environmental, economic and physical requirements and at
the same time, ensures the involvement of different stakeholders. Better land-use evaluation
20
helps to prepare more sustainable land-use plan. But, fitting different social, economic,
environmental, physical and stakeholders’ requirements during land-use evaluation is a tricky
job. It’s a complex decision making process. Different Multi-Criteria Decision Making
(MCDM) techniques can be useful to ease such complex decisions. So, from section-4.3
some brief discussion on MCDM has been highlighted.
4.2 Land-use Classification
Harland Bartholomew approached the first land-use classification technique, where he
applied a two-level land-use classification system (please see Table-5) (Lovelace 1993). In
his land-use categories he considered activity based classification, like, residential,
commercial and industrial; designated public uses; and separated vacant or undeveloped land
from others. In the second level he further classified some first level land-uses in terms of
intensity of activities and types of activities. Thus, Bartholomew developed the first land-use
classification system that helped to build a standard form for data collection and comparisons
for land-use analysis.
Table 5: Bartholomew land-use classification (Lovelace 1993)
Level 1 Level 2
Residential Single-family homes
Two-family homes
Multiple dwellings (apartments)
Commercial
Industrial Light industry
Heavy industry
Public and semi-public Schools, churches, hospitals, institutions, golf
courses, etc.
Public parks
Railroads
Streets
Vacant land Undeveloped or agricultural
Bartholomew classified land-use based on USA cities. Later, due to improvement of
technologies in surveying, data collections and data management systems, more modern and
21
detail land-use classification techniques evolved. In this evolutionary process the most cited
documents on land-use classification systems can be also found from USA context.
Urban Renewal Administration of USA et al. (1965) introduced a code wise land-use
classification using different level of numeric digits like, one-digit, two-digit, three-digit and
so on to represent a particular land class. They published a Standard Land-Use Coding
Manual (SLUCM).
Due to advent of remote sensing images, Anderson (1976) suggested a four level land-use
classification system. At the first level, land-use categories were urban or built-up land;
agricultural land; rangeland; forest land; water; wetland; barren land; tundra; and perennial
snow or ice. This classification is based on land-cover, which, can be interpreted by using
remote sensing images. The first level land-use classification can be used for a large scale
area, like nationwide. The second, third and fourth level classification permits much
flexibility to adjust with local level needs, like city planning.
In 1994, the Research Department of the American Planning Association assisted Federal
agencies to set a Land-Based Classification Standards (LBCS), which was an updated version
of 1965 SLCUM (Jeer 2001). LBCS is a land-use classification model, which consists five
dimensions and each dimension has different levels. The dimensions are activities, functions,
building types, site development character, and ownership. In briefly, the dimensions can be
explained by the following ways (Wang and Hofe 2007) –
• Activity dimension – it directly describes human use of land, like, residential,
shopping, business or trade, travel, leisure, natural resources etc.
• Functional dimension – it reflects economic function rather than actual activity. For
example, the activity of an establishment is shopping, but, it falls under the retail sales
and service category according to the functional dimension. Besides, when one
establishment shows multiple activities, but, has single function, then functional
dimension is used to classify land. For example, one establishment may have different
activities like offices and factory, but, as it has single economic function, so, it falls
under a particular functional category, like as textiles.
• Structure dimension or building type – it simply reflects the structure types on land,
like, residential building, commercial building, school building etc.
22
• Site development dimension – it refers to physical character of the land. Such
classification identifies which land is developed and which is not, which land is
occupied by building structures and which is not, etc.
• Ownership dimension – as it is named, it reflects the ownership of the land. This
dimension of land-use classification shows either a land is owned publicly, privately,
or non-profitably etc.
Thus using multi dimension classification techniques user can control the precision of land-
use categories at different level. The LBCS supports planning applications at different
geographical scales like neighborhood scale, city scale, regional scale or national scale.
Bangladesh followed a paradigm shift in the urban land-use classification during the Detail
Area Plan (DAP) phase of Dhaka Metropolitan Development Plan (DMDP) project. It was
one of the early approaches on GIS based urban planning in context of Bangladesh. The
improved data acquisition technology and application of GIS tools allowed planners to adopt
a wide and extensive level of land-use classification technique. Recently, the planning
authority, Rajdhani Unnoyon Kartipakkha (RAJUK), area trying to make further
improvement of existing land-use classification system. Some Planner Consultants of
RAJUK proposed a three level land-use classification system. The first level is the general
land-use classes. They proposed 20 general land-use categories. Under these general
categories they proposed more detail land-use classes at 2nd level and most detail land-uses at
the 3rd level, which is under 2nd level land classes. In their proposed classification system,
they tried to address the issues they faced during the implementation of DAP. Such land-uses
will help the planners during the plan preparation as well as plan administration.
4.3 Criteria for Urban Land-use Planning
There are no unique factors and criteria for urban land-use planning. Dai, Lee and Zhang
(2001) applied comparisons of topography, ground conditions, grandwater and geologic
hzards for land-use planning of Lanzhou city, where, they made geo-environmental
evaluation. Dong, J. et al. (2008) claimed an integrated evaluation of urban development
suitability in their study. They considered 11 paparameters under the factors of environmental
backgrounds, water/land resources and socio-economic development. But their parameters
don’t look as complete for an integrated evaluation. They neighter considered community
services like schools, hospitals etc. nor they imposed utility services like water supply,
electricity supply etc. in their evaluation. Tudes and Yigiter (2010) proposed land-use
23
planning for high storey buildings, multi storey buildings, low storey buldings, industrial
sites, waste disposal sites and green lands in their study in Adana-Turkey. They used varying
conditions of environmental criteria, which are slope, elevation, surface geology, depth of
ground water, bearing capacity, agricutural suitability, land-use, earthquake susceptability
and flooding area. Their selcetion of criteria was rather complete according to their proposed
types of land-use. So, from the selected literatures it was found that there is unified rule for
urban land-use planning. Researchers identified their own criteria of land-use planning
according to their subjective judgment, theoretical knowledge as well as local in-situ
knowledge.
4.4 Multi-Criteria Decision Making Model (MCDM) and GIS
Several MCDM techniques have been evolved to solve complexity in human decisions. MCD
techniques help decision makers to consider wide level of information and assist to deal with
complexity of information in a convenient way. Using MCDM techniques decision makers
can combine a lot of information, can make comparative analysis and can choose best
alternative from a number of options. MCDM techniques guide decision makers to think
consistently, especially, when they become puzzled with the elements of decisions.
There are many MCDM techniques and their number is still rising. The variations in MCDM
approaches occur due to several reasons like, types of decision, time availability, amount or
nature of data, the analytical skills of those supporting the decision, administrative culture
and requirements of organizations etc. (Department for Communities and Local Government:
London 2009). However, from literature reviews two basic types of MCDM techniques can
be distinguished – multiple attribute decision making (MADM) and multiple objective
decision making (MODM) (Kahraman 2008). MADM identifies best alternative from a set of
finite number of alternative solutions, while, MODM deals with infinitive number of
alternatives.
Traditionally GIS based land suitability analysis was done using simple overlay method
(Malczewski 2004). For the last two decades, integration of MCDM techniques with GIS has
made considerable improvement in land suitability analysis. MCDM techniques allow to set
decision making rules, criteria, intensity or level of given preferences as well as to set
functional relationships among the rules and criteria by using their methodological
operations. On the other hand, GIS has huge capabilities for spatial and non-spatial data
acquisition, storage, retrieval, manipulation and analysis. So, MCDM techniques embedded
24
with GIS can be used fruitfully for land suitability analysis as well as for land-use planning as
such planning approaches need involvement of different decision rules, analysis of wide level
of spatial and non-spatial data and need combining all the decision making factors to get the
results. GIS provides an output map of land suitability analysis by combining all the decision
inputs in the realm of a particular MCDM technique.
A number of MCDM techniques embedded with GIS have been applied for land suitability
analysis (Malczewski 2004). Some of them are MODM techniques and some are MADM
techniques. MODM techniques, like heuristic approaches, AI techniques, require
mathematical programming algorithms which are much complicated. While, there are several
MADM techniques, which are easy to understand and much useful for land suitability
analysis, especially, in raster environment. Such techniques are Weighted Linear
Combinations (WLC) (or simple additive weighting), ideal point methods, concordance
analysis, Analytical Hierarchy Process (AHP), Fuzzy Logic, Ordered Weighted Averaging
(OWA) etc. Among those techniques, from the literature survey it was found AHP is most
widely used technique for urban land-use planning (Dai, Lee and Zhang 2001; Zhang, Li and
Fung 2012; Tudes and Yigiter 2010; Dong, J. et al. 2008).
There are several problems in applying MCDM techniques in GIS environment (Malczewski
2004). First, the input data may have property of inaccuracy, imprecision, and ambiguity.
Some MCDM techniques need more accurate data to make judgment more objectively, while,
some MCDM techniques can deal with less accurate data by making subjective judgment.
Second, the problem associates with standardization of criteria. Different methods
standardize criteria in different ways that shows different land-use suitability.
Standardizations of criteria in some techniques are complicated and for some techniques it is
simple. All standardization methods are not suitable for a particular study. For example, for
the soil fertility analysis Fuzzy Logic can be appropriate, as the merits and demerits of the
properties of soil should be determined in many-values but the Boolean values. Third, the
problem is finding a proper justification on which method is more suitable. This problem
associates with the first and second problems. An MCDM technique for a particular study can
be supported in consideration of second problem, but, cannot be supported for the first
problem. So, to minimize such problems, specific MCDM techniques should be selected
based on data availability, technical expertise, time availability etc.
25
4.5 The Analytical Hierarchy Process (AHP)
AHP is one of the most popular MCDM techniques developed by Saaty (1980). It is used to
identify the best one from a set of alternatives with respect to several criteria. The basic
principle of AHP is to solve a problem by forming hierarchies. At first, a hierarchy tree is
drawn to segregate each criterion into lower sub-criteria at two or more levels. At the top of
hierarchy tree, the goal is set, and then, major criteria and sub-criteria are fixed. Thus AHP
helps to make assessment from lower criteria; each criteria and sub-criteria have individual
performance to achieve the goal. Following are the steps performed for GIS-AHP based land
suitability analysis (Ullah and Hafiz 2014) -
(i) Making pair-wise comparison among criteria,
(ii) Preparing comparison matrices,
(iii) Standardization of matrix values,
(iv) Checking consistency ration and finalizing the relative weight values,
(v) Transforming the weight values into spatial database,
(vi) Overlay operations and preparing composite map,
(vii) Classifying the composite map into highly suitable, moderately suitable and marginally
suitable area.
The detail steps of AHP have been described in the following chapter (Chapter 5).
4.6 Why AHP has been used for this study
Different social, economic, environmental and physical factors and criteria influence the
land-use planning decision. It is a complex job to categorize the criteria because of conflicts
among criteria and for their degree of effects on land-use planning. For example, both
agricultural land and parks can be defined under open space category. But, in that case,
agriculture can be treated suitable for recreational development that may have negative
impacts due to scarcity of agricultural land and for food deficit. So, it is necessary to
categorize the criteria very sensitively. But, it becomes hazy to categorize the criteria in a
structured way. AHP allow to set factors and criteria in a structured way. Features under this
criteria and sub-criteria may exist in separate locations or can coexist in a single location. So,
each criterion needs to evaluate individually and it needs to sum up all the criteria on a given
location for estimating its level of preference. AHP embedded with GIS can easily handle the
matter with their measurement and overlay techniques. For land-use planning, some decision
rules are necessary to be applied. Sometimes those decision rules can be fixed objectively by
using quantitative measurements and sometimes it is necessary to apply subjective judgment
26
to set the decision making rules. AHP can handle both the objective and subjective judgment
as well. For imposing subjective judgments, decision-makers sometimes fail to maintain
consistency. AHP guides to make consistent judgment. It helps to minimize contradictions
among a set of criteria. For sustainable land-use planning, there are necessary to include
different experts’ opinions as well as that of non-expert stakeholders. Then, it becomes a
tricky job to emphasize whose opinions are more viable. AHP make space for all individuals
or groups from different backgrounds. For a larger area land-use suitability should be ranked
in priority order. For example, an area, which is in highland and is well connected by civic
facilities, should have high priority for residential development. AHP embedded with GIS
can rank the suitable areas in different order like highly suitable, moderately suitable, low
suitable or not suitable areas.
27
Chapter 5: Designing Geodatabase and
Evaluation of Land Suitability using AHP
This chapter focuses major methodological operations of the thesis works. Through literature
reviews and by sharing experts’ opinions, the requirement analysis was done; the geodatabase
was designed as well as the criteria influencing urban land-use planning were selected. After
processing the collected data, land suitability was evaluated using AHP. For this entire
methodological operation, it was kept in mind that the results would be used for urban land-
use planning and its administration at a regional scale in view to promote sustainable urban
development. It would not be a detail land-use plan but would be useful for a strategic
planning like DMDP Structure Plan and would minimize the time and resources used for
preparing DAP.
In context of Dhaka city, the priority area of urban development should be residential
development, industrial and commercial development, recreational development and at the
same time protecting the agricultural land. As a one of the fastest growing megacities in the
world, the immediate necessity of Dhaka city is providing adequate shelters in the form of
residential development. As a growing economy the primate city of the country, Dhaka needs
room for commercial and industrial development. The current unplanned development of
Dhaka city is limiting the parks, open spaces and agricultural land. So, it is also important to
create parks, open spaces and other outdoor recreational facilities, and protecting the
agricultural land for food sustainability. In this view, the evaluation process of land suitability
of this thesis selected the criteria that affect those development issues. The priority of the
criteria was set in a way so that higher suitable area could be proposed for hard urban
development like residential, commercial and industrial development and lower suitable area
could be reserved for agriculture, park, open spaces and other recreational facilities. In the
next chapter, more discussion was made to highlight how priority orders of the study area
were analyzed for proposing different land-uses.
5.1 Modeling Geodatabase using UML
The UML-Geoframe concept, described in chapter 3, was applied for designing geodatabase
model of urban land-use planning. Two objectives were set for this conceptual modeling –
• Fixing data requirements and database structure to prepare land-use plan for DMDP
(Dhaka Metropolitan Development Plan) area.
28
• Handover the datasets to the authority (RAJUK, the capital improvement authority,
who is mainly responsible for planning, managing and controlling of land-use within
DMDP area) to administer the land-use plan.
The broad themes for land-use planning and management under DMDP area are shown as in
Figure 12:
Figure 12: Broad themes of land-use planning
5.1.1. Administrative and Planning Boundary
This theme will mainly require for RAJUK for administrative and management purposes after
finishing the land-use plan. The land-use plan will set guidelines for administration of land-
use. Following information was step-by-step followed for preparing class diagram under this
theme (Figure – 13)–
• DMDP area consists of several districts (fully or partly). Total Bangladesh is divided
by 64 districts.
• Each district comprises number of Thana (sub-district).
• The major built-up portion of DMDP area is Dhaka City Corporation, which consists
of 14 Thana. Recently, Dhaka City Corporation has been divided by two parts –
Dhaka City Corporation North and Dhaka City Corporation South. So, name field has
been added.
• Dhaka City Corporation is divided by around 100 wards. The name of the ward class
of Dhaka City Corporation has been termed as Ward_C.
• There are other Pourashava towns (municipalities) in DMDP. Generally, a Pourashava
exists within a single Thana. But in some cases, a purashava can overlap to another
Thana. So, here aggregation relationship was used.
Administrative and Planning Boundary
Land-use
Dhaka Metropolitan Area
Physical and Geological
Utilities
Community Facilities
Demographic
29
• Each Pourashava is divided by 9 wards. The name of ward class of Pourashava is
termed as Ward_P.
• Besides Pourashava area, other areas of a Thana belong to Union Parishad (the rural
local government institute). There are number of unions under each Thana.
• Each union contains 9 wards. But, the ward boundaries under union are not properly
delineated in spatially. However, it will not make any problem to identify a parcel of
land without considering the ward boundary for union areas, which are predominantly
rural. In those areas, it will be better to use Mouza name. So, ward boundary in union
areas were not considered for database.
• Mouza is the unit for keeping land record and is administered by Land Record office
as well as by Land Ministry. Alternatively, it can be said as cadastral unit.
• Each union consists of one or more mouza, but, there may have overlapping. One
mouza can be extended between two or more unions. So, aggregation relationship
was used between union and mouza.
• Each mouza contains one or more blocks, which are delineated by Mouza Sheet.
• Each Mouza sheet has number of plots, that is, the ultimate parcel of land.
• It is very much important to know the plot ownership pattern for planners and
authority. So, an enumeration class of plot ownership pattern has been added.
• Ward boundary of City Corporation and Pourashavas can be properly delineated by
plot no. So, here composition relationship was used.
• The total DMDP area was divided into several group boundaries to prepare the detail
land-use plan.
• Each groups were again divided by several number of SPZ (special planning zone)
boundaries.
• During the plan preparation, planners divided each group into several DAP (Detail
Area Planning) zone.
• Such grouping and zoning measures are taken during the planning phase. After
finishing the plan, the land-use is administered by identifying plot no., ward no. etc.
Such group or zoning boundary is not necessary for day to day work of authority. So,
the group boundaries were presented in separate class diagram without making any
link with land parcel and other administrative information.
30
Figure 13: Class Diagram - Administrative and Planning Boundary theme
31
5.1.2 Land-use
During designing class diagram for land-use theme following points were considered (Figure
- 14) -
• Planners suggested preparing land-use in three hierarchy/categories: General land-use,
Intermediate land-use and Detailed land-use.
• The primary data is procured from detailed land-use.
• The structure data is also used for preparing detailed land-use. For example, a
building is a high school. So, a land-use of school is derived from that building.
• Structure may be one storied or multi-storied. So, it is necessary to know the no. of
floor.
• Structures are in three type – Pucca (Hard brick, cement or similar structure), Katcha
(soft straw, tin or similar structure) and Semi-pucca (Brick/Cement with tin/straw
etc.).
• Land-use type can be existing or proposed in similar manner. So, Land-use_type
attribute was added under Land-use_Detailed.
• Some environmental laws (like wet land act) control certain land-uses for particular
areas.
Figure 14: Class Diagram – Land-use theme
32
5.1.3 Physical and Geological
The main considerations for Physical and Geological theme were as follows (Figure 15) –
• Elevation, soil quality and distance to fault line are the datasets require under physical
and geological theme. Because such features help to study the feasibility of
construction sites, landing built-up areas, evaluation of earthquake risks and other
geo-environmental phenomenon.
• Ground water condition was not considered as already experts are suggesting for
collecting water from subsurface source for Dhaka city.
-Rl_m : esriFieldTypeDouble
SpotHeightContour
-Soil_Type : SoilType
Soil+Rock
+Hard
+Soft
«enumeration»
SoilTypeFaultLine
+Shape : esriFieldTypeGeometry
ESRI Classes::Feature
+OBJECTID : esriFieldTypeOID
ESRI Classes::Object
Figure 15: Class Diagram – Physical and Geological theme
5.1.4 Utilities
The principle considerations of Utilities theme are (Figure-16) –
• The major utilities services that have influences on land-use planning are road, drain,
water supply, gas supply, electric supply and sewerage facilities.
• Road types are an important factor. Road types may be highway, pucca (metal
road/carpeting road), semi-pucca (brick-built road), katcha (mud road), footpath,
embankment and railway. The context of road hierarchy, like primary road, secondary
road etc., is not usually applicable in perspective of Bangladesh. Specially, in the
study area such hierarchy could not be maintained.
33
• Waterway has not been considered. Because, waterways have little impact over land-
use of Dhaka city and a waterbus service project along the city failed. The feasibility
of waterway services should be studied under transportation planning.
Figure 16: Class Diagram – Utilities theme
5.1.5 Community Facilities
Designing class diagram for community facilities was very simple as only point data of
certain objects, like school, healthcare center, park, bus stop and market/bazar, were major
considerations (Figure-17).
Figure 17: Class Diagram – Community Facilities theme
34
5.2 Data Processing
Different sets of secondary data were collected for this study (see section – 2.2). Data on
administrative theme were not collected for this study. Because, this theme has no influence
on proposed land-use planning. However, the requirements of the theme were designed in
UML as this theme is important for planning authority to manage a land-use plan. Most of the
data were collected from the source of RAKUK Detail Area Plan survey. In the study area,
there is no service of Water Supply Authority, which is also responsible for providing
drainage and sewerage facilities. So, no datasets were collected for those facilities. Data were
also unavailable for soil types, which is much significant to analyze earthquake risks. But,
the fault lines passing through the major rivers were used to determine the zones of
earthquake risk (see section – 2.5).
All the data collected was in Esri Shape file format except data on Gas Supply. Gas Supply
data was in paper map format which was digitized to store in GIS platform. The entire shape
files were exported into an Esri Geodatabase.
An important task should be to process features in geodatabase so that the features maintain
accurate topological relationships. There should not have overlapping in features, sliver
polygons and associated problems. That is, efforts are necessary to ensure quality database. In
this thesis, such jobs were avoided as that would be huge time consuming. Moreover, such
job would be just a technical work rather than making something new inventory for research.
5.3 Selection of Criteria
In AHP process selection of criteria and their sub-criteria is a crucial stage. Because,
selection of criteria influence the judgment by segregating one criteria from other and at the
same time, by giving more importance to one criteria over other. By synthesizing numbers of
literature reviews, local contexts and expert opinions a number of factor and criteria have
been selected for this study. The criteria were further divided into several factor ratings
(Classes) depending on their importance or preferences at different levels. Except the land-
use criteria, all the criteria were classified into different factor ratings based on distances.
Land-use is not a point or line feature, rather an area feature. Here, classification based on
distances is not applicable. So, no classification was made for land-use criteria. The factors,
criteria and their ratings can be shown in the following Table-6 –
35
Table 6: Factors and Criteria of Land-use
Factors Criteria Unit Factor Rating
Class-1 Class-2 Class-3 Not
suit.
Land-use Agricultural Type Agricultural
Homestead Type Homestead
Residential Type Residential
Vacant Land Type Vacant Land
Utilities Proximity to
Road
Meter 250 500 750 >750
Proximity to
Gas Supply
Meter 250 500 750 >750
Proximity to
Electricity line
Meter 250 500 750 >750
Physical &
Geological
Elevation Meter 3-6 6-9 9-13
Distance from
Fault Line
Meter >250 <250
Community
Facilities
Distance from
School
Meter 250 500 750 >750
Proximity to
Healthcare
centre
Meter 500 1000 2000 >2000
Proximity to
Bus Stop
Meter 250 500 750 >750
Proximity to
Market/Bazar
Meter 250 500 750 >750
Proximity to
Park*
Kilometer 2 5 10 >10
*Reserved Open space has been considered as Park as there is no Park services data. This
open space supposed to can be used as park.
5.4 Preparing Comparison Matrix
A useful step of AHP is making Comparison Matrix. The comparison matrix is prepared from
Pair-wise Comparison. A Pair-wise Comparison, suppose comparison on how important is
the A than the B, is performed in 9 degree preferences scale as suggested by Saaty (1980). At
each higher level of scale shows higher importance than the previous lower level (Table-7).
36
Although Saaty (1980) suggested his 9 degree preferences scale for qualitative judgment
based on experiments but at the same time his stance was flexible. Other suitable scale can be
followed. Many authors criticized Saaty’s numeric scale and several authors tried to improve
that scale in alternative ways, but still no unique scale has been suggested other than Saaty’s
scale (Harker and Vergas 1987; Saaty 1990; Lootsma 1993; Dong, Y. et al. 2008). On the
other hand, Saaty’s ratio scale is easy to understand for decision makers and researchers from
wide level of backgrounds are using this scale successfully to ease conflict decision making
process.
Table 7: Fundamental scale used in Pair-wise Comparison (Saaty and Vargas 2001)
Intensity of
Importance
Qualitative Definition Explanation
1 Equal importance Two activities contribute equally to the objective
2 Weak
3 Moderate importance Experience and judgments slightly favor one
activity over another
4 Moderate plus
5 Strong importance Experience and judgment strongly favor one
activity over another
6 Strong plus
7 Very strong or demonstrated
importance
An activity is favored very strongly over another
and dominance is demonstrated in practice
8 Very, very strong
9 Extreme importance The evidence favoring one activity over another is
of the highest possible order of affirmation
To make the pair-wise comparison between two factors or criteria under 9 degree preferences
scale following Figure-18 was used –
Land-use
9 8 7 6 5 4 3 2 1 2 3 4 5 6 7 8 9 Physical/Geological
Figure 18: Diagram to choose preference level between Land-use and Physical/Geological.
Using Figure-18, at first an expert fixes his stance, either both criteria are equally important
or not. If they are equally important, then the value is 1. If they are not, then the expert takes
his position where he prefers. For example in Figure-18, an expert prefers Land-use to
Physical/Geological. So, he takes left side positions of 1. Finally, according to 9 degree
37
preferences scale (Table-3) he marks his actual value of preference. In this diagram, the land-
use got a weak favor over the factor of physical/geological. So, the value of 2 at the left side
of 1 was highlighted. Thus pair-wise comparisons were made for all the factors. The number
of pair-wise comparison can be calculated using the following formula (Saaty and Vargas
2001) –
n(n – 1)
2
Where, n is number of total criteria or factors. Suppose, we have 4 factors in our study. So,
the numbers of pair-wise comparison were 6 at the first level.
After making pair-wise comparison, Comparison Matrix is prepared. For example, a 4 by 4
matrix was prepared for the 4 factors. The diagonal values of the matrix are always 1. We fill
up the upper triangular matrix at first. If the values are at the left side of 1 (like Figure-18),
we put actual judgment. If it is at the right side of 1 then we put reciprocal value.
After filling the upper triangle we fill the lower triangle with the reciprocal values. The
formula is simple. If aij is the element of row i column j of the matrix, then the lower
diagonal is
aij =
So, we get one factor matrix and four criteria matrices from each expert (see Appendix - 1).
5.5 Estimating Weight for Factors and Criteria
For estimating weight for factors and criteria, at first level and second level, following steps
were followed (Saaty and Vargas 2001) –
• Summing each column of comparison matrixes prepared in previous section (section –
5.4). For the example, at first level, we get from an expert –
Land-use Physical/
Geological
Utilities Community
Facilities
Land-use 1 2 7 8
Physical/Geological 1/2 1 6 7
Utilities 1/7 1/6 1 2
Community Facilities 1/8 1/7 1/2 1
Sum 1.77 3.31 14.5 18
Equation (1)
Equation (2) 1 aji
38
• Dividing each element of matrixes with the sum of their columns. Here we get
normalized relative weight, where, the sum of each column is 1.
Land-use Physical/
Geological
Utilities Community
Facilities
Land-use 0.56 0.60 0.48 0.44
Physical/Geological 0.28 0.30 0.41 0.39
Utilities 0.08 0.05 0.06 0.11
Community Facilities 0.07 0.04 0.03 0.55
Sum 1 1 1 1
• The Normalized Principal Eigen Vector was obtained by averaging across the rows –
W = 1/4
The normalized principle Eigen Vector is also called Priority Vector. As this is normalized,
so, the sum of elements of priority vector is 1. Each element of priority vector shows the
relative weight of its corresponding criteria. For example, the relative weight of land-use is
52%. But, before finalizing this relative weight, consistency of preferences was checked. The
consistency is checked by following some steps of set formulas.
• To check the consistency, at first Principle Eigen Value was calculated. The calculation
was obtained from the summation of products between each element of Priority vector
and the sum of columns of the comparison matrix.
λmax = 1.77(0.52) + 3.31(0.35) + 14.50(0.08) + 18(0.05) = 4.12
Then Consistency Index (CI) was calculated using Equation-3. The CI is used to find the
deviation or degree of consistency. But, this equation is not final. To do the final check
Saaty(1980) proposed Consistency Ratio (CR), which is a comparison between CI and
Random Consistency Index (RI), the appropriate one (Equation-4). Saaty (1980) determined
0.56 + 0.60 + 0.48 + 0.44 0.52
0.28 + 0.30 + 0.41 + 0.39 0.35
0.08 + 0.05 + 0.06 + 0.11 = 0.08
0.07 + 0.04 + 0.03 + 0.55 0.05
39
standard RI against different number of criteria from 500 sample matrices. For example, RI is
0.9 for 4 numbers of criteria. From Equation – 3 & 4, we get CI = 0.04 & CR = 0.04 for this
four factors example. Saaty (1980) suggested if the CR is less than or equal to 10%, then, the
inconsistency is acceptable. In our all cases of study (both at the first level and second level),
the CR values were remained below 10%.
CI =
CR =
To do the whole weight estimation job, the calculation formulas were designed in MS Excel
program. During sharing with expert opinion, experts were asked to give their preferences
values on a 9 degree scale (like Figure-18). Based upon their judgment, if the values of CR
were shown as greater than 10%, then experts were asked to repeat their task with different
preferences. Thus, the weight assessments from individual experts have been shown in
Appendix - 2.
A major challenge here was to aggregate different preferences values of experts into a single
rating value. It was not possible to call all the experts for a group discussion and finalize the
single ratings by them. In this connection, a number of literatures were reviewed. Several
approaches have been suggested to aggregate individual judgments or priorities (Ramanathan
and Ganesh 1994; Ernest and Kirti 1998; Dong et al. 2010; Aczel and Satty 1983). The main
objective of those approaches is to reach at the Pareto (unanimity, agreement) Principle. In
this study the Geometric Mean of individual priority was considered. That is, the geometric
mean of all individual experts’ Principle Eigen Vectors of a particular criterion was
calculated. Then, the geometric means of all criteria were normalized by dividing with their
sum so that disparity can be minimized (see Appendix-2).
Thus we get final weight of all the factors and criteria at the first and second level. At the end
of this stage, we get the overall weight by multiplying the weight gotten in level one with that
in level two (Table – 8).
Equation (3)
Equation (4)
λmax – n
n - 1
CI
RI
40
Table 8: Weight values of factors, criteria and factor ratings
Level 1 Level 2 Overall Weight Rating (Level 3) Factor W1 Criteria W2 (Wi=W1xW2) Class-1 Class-2 Class-3 Not Suit. Land-use 0.50 Agriculture 0.10 0.049 - - - - Homestead 0.39 0.195 - - - - Residential 0.07 0.036 - - - - Vacant land 0.44 0.219 - - - - Physical Geological 0.34 Elevation 0.40 0.135 0.125 0.313 0.563 0 Fault line 0.60 0.205 1 - - 0 Utilities 0.10 Road 0.67 0.067 0.562 0.312 0.125 0 Gas Supply 0.07 0.007 0.562 0.312 0.125 0 Electricity 0.26 0.026 0.562 0.312 0.125 0 Community Characteristics 0.05 School 0.52 0.026 0.562 0.312 0.125 0 Healthcare Centre 0.22 0.011 0.562 0.312 0.125 0 Bus Stop 0.09 0.004 0.562 0.312 0.125 0 Market/Bazar 0.12 0.006 0.562 0.312 0.125 0 Park 0.05 0.003 0.562 0.312 0.125 0
41
At the third level, there is no overlapping condition. So, here comparison matrices are not
necessary. But at different unit of intervals different level of scores were assigned. The scores
were determined by the researcher’s own judgment as he already have perceived idea about
each criteria during the literature reviews and sharing experts’ opinion survey. The score
values were fixed according to AHP 9 degree scale. As this scale was already used in first
and second level, so, it seemed convenient to use the same scale at the third level. Then, the
score values were normalized to minimize disparity among the values. At an interval unit
where a particular criterion is not suit, there zero value was assigned. In case of land-use
criteria, there is no overlapping condition in the second level. So, for these criteria no score
values were assigned rather the overall weights gotten from first and second level were used
(Table - 8).
5.6 Evaluation of Land Suitability in GIS
To evaluate land suitability in GIS, the spatial database was prepared with data layers. Each
data layer represents a particular criterion. The data layers were constructed from buffer
analysis at the given distance units for each criteria. The attribute of these data layers were
determined by multiplying the normalized score values at third level with the overall weights.
Then, each data layers were converted to raster format where the attribute values were used.
That is, each criterion values are represented by a particular raster data layer. The raster
operations were performed on 300m grid cells.
But, it was not necessary to buffering the land-use criteria as those criteria were already in
polygon and there was no need of multiplication. On the other hand, elevation data was taken
by creating Digital Elevation Model (DEM), which is also already in raster format. DEM was
generated in Inverse Distance Weight (IDW) method as there was low resolution spot height
data. DEM values were reclassified to prepare criterion map in raster format.
After creating all the required raster data layers, they were overlaid or aggregated in raster
calculator. Thus a composite map was produced. The values of composite map show the
land-use suitability index. The entire calculation at this stage can be shown by the following
equation –
n S= ΣWi*Xi
i=1 Equation (5)
42
Where, S is the land-use suitability index. Wi are the overall weights which were gotten
previously by multiplying the weights of first level and second level criteria. Xi scores are
normalized value of lowest level factors.
But, still one thing should be considered for evaluating suitability index. Some areas are
restricted for existing university campus, army camps and for other reasons. These areas
should be excluded from land-use planning. To do that job a Boolean raster should be
multiplied with the suitability index, where the value of restricted areas in Boolean raster is 0
and other areas are 1. This can be simplified by the following equation –
Where, π is the Boolean value.
Finally, the suitability index was classified into four classes to rank the study area into highly
suitable, moderately suitable, low suitable and not suitable locations (Figure-19). As most of
the criteria were rated under four classes so the final composite map were also classified into
four categories. This suitability map was thereafter used for land-use planning.
n
S= ΣWi*Xi*π i=1
Equation (6)
43
Figure 19: Land-use Suitability Index map
44
Chapter 6: Result, Discussion and
Conclusion
6.1 Result and Discussions
6.1.1. The Land-use Suitability Indexes
As stated in the previous chapter, the suitability indexes were classified into four classes –
highly suitable, moderately suitable, low suitable and not suitable.
Table 9: Area of different suitability types
Suitability Type No. of Grid Cells Area (in sq. km) Proportion of Area
Highly Suitable 436 39.24 13%
Moderately Suitable 1186 106.74 35%
Low Suitable 1438 129.42 42%
Not Suitable 346 31.14 10%
Total 3406 306.54 100%
Table-9 shows that only 39.24 sq. km area is highly suitable (13%) followed by 106.74 sq.
km is moderately (35%) and 129.42 sq. km (42%) is low suitable area. If we overlay each
criterion map on this suitability index map, then, we get following findings –
6.1.1.1 Land-uses
Table 10: Land-use classes under each suitability types
Land-use Type Suitability Type
Agriculture Homestead Residential Vacant Land
Area (%) Area (%) Area (%) Area (%) Highly Suitable 0 0% 0 0% 0.27 9% 38.97 37% Moderately Suitable 22.77 16% 37.26 61% 2.43 82% 44.28 42% Low Suitable 90.27 65% 20.52 34% 0.09 3% 18.54 18% Not Suitable 25.38 18% 3.15 5% 0.18 6% 2.43 2% Total 138.42 100% 60.93 100% 2.97 100% 104.22 100%
Table-10 shows that most of the agricultural land falls into low suitable and not suitable area
(65% & 18%). Around 16% of total agricultural land belongs to moderately suitable area,
while, no agricultural land was found in highly suitable area. This happened, because, experts
preferred to avoid agricultural land from residential, commercial and similar hard urban
development. A substantial portion of homestead, residential and vacant land (61%, 82% &
45
42%) falls into the category of moderately suitable land. These types of land-uses are suitable
for such hard urban development. Some 9% of residential land and 37% of vacant land are
located within highly suitable area. This two types of land can get priority for urban
residential development, because - The residential land-uses are already in residential. So, the
existing residential land-uses could be promoted into planned residential area to facilitate
more housing. Then, new housing development also can be promoted in the vacant lands.
Figure 20: Comparison between suitability types and land-use criteria
6.1.1.2 Physical and Geological
Table 11: Elevation classes under each suitability types
Elevation
Suitability Type
3 to 6 m 6 to 9 m 9 to 13 m
Area (%) Area (%) Area (%)
Highly Suitable 2.07 2% 12.24 10% 16.83 20%
Moderately Suitable 36.63 41% 48.06 40% 44.73 52%
Low Suitable 34.38 39% 52.47 44% 19.89 23%
Not Suitable 15.75 18% 19.44 16% 4.05 5%
Total 86.76 100% 119.97 100% 85.5 100%
46
Table-11 illustrates that 16.83 sq.km (20%) of 9 to 13 meter elevated areas falls into highly
suitable area. This portion of area is comparatively higher than the portions of lower elevated
area (2% and 10%) in the same suitability category. Such figure justify that more than 50% of
highly suitable land fall into high land (9 to 13 meter), which is free from flood and favorable
for residential development. Most of the significant portion of all types of elevation go under
moderately and low suitable area. The facts for this significance is that the sizes of the
moderately suitable and low suitable area are also significantly higher (35% and 42%, see
Table -9) and there are little variation of elevation in the study area, meaning that the study
area is primarily a plain land.
Figure 21: Comparison between suitability types and elevation criteria
47
Table 12: Earthquake risk area under each suitability types
Fault line
Suitability Type
Within Risk Not in Risk
Area (%) Area (%) Highly Suitable 0 0% 39.24 16% Moderately Suitable 4.41 8% 102.33 41% Low Suitable 22.68 41% 106.74 42% Not Suitable 28.35 51% 2.79 1% Total 55.44 100% 251.1 100%
Table-12 depicts that highly suitable area has no risk of earthquake. So, residential land-use
can be proposed in this area as people usually stay at their resident at night. Only 8% of risky
area is under moderately suitable area. This marginal portion of land could be used for urban
development other than residential, commercial and industrial development. Around 92% of
risky area falls within low suitable and not suitable area. So, the result validates that low
suitable and not suitable area should be protected from hard urban development.
Figure 22: Comparison between suitability types and fault line criteria
48
6.1.1.3 Utilities
Table 13: Road classes under each suitability types
Proximity to Road Suitability Type
Within 250 m Within 500 m Within 750 m Area (%) Area (%) Area (%)
Highly Suitable 39.24 36% 0 0% 0 0% Moderately Suitable 64.44 59% 30.15 27% 12.15 14% Low Suitable 4.05 4% 71.91 65% 53.46 62% Not Suitable 1.26 1% 9 8% 20.88 24% Total 69.75 100% 111.06 100% 86.49 100%
According to Table- 13 and Table-9, all the highly suitable area (39.24 sq. km) lies within the
250 meter buffer zone of road networks, that is, highly suitable are is well connected by road
networks. Similarly, moderately suitable area is also close to the road network as 59% of the
area within the 250m distance of roads fall under moderately suitable area. On the other hand,
low suitable area and not suitable area located at the far distances from the roads. So, it can
be inferred that land-uses, like residential, commercial and industrial developments, can be
proposed in highly and moderately suitable areas. An almost similar result (with almost a
similar figures) is also found in case of electricity (Table-14), because, electricity lines passes
through the road networks. So, similar urban developments also can be proposed in
consideration of electricity. Although the figures for gas line (Table- 15) are different, but the
summary of the result is same – highly and moderately suitable areas are close to gas line,
while, the low suitable and not suitable areas are in far distances. So, the land-use proposal in
connection to the criteria of gas line should be similar to other utility services like road
networks and electricity.
Table 14: Electricity classes under each suitability types
Proximity to Electricity
Suitability Type Within 250 m Within 500 m Within 750 m
Above 750 m (not Suit.)
Area (%) Area (%) Area (%) Area (%) Highly Suitable 39.24 36% Moderately Suitable 64.44 59% 20.61 31% 9.54 21% 12.15 14% Low Suitable 4.05 4% 41.22 62% 30.7 69% 53.46 62% Not Suitable 1.26 1% 4.68 7% 4.32 10% 20.88 24% Total 69.75 100% 66.51 100% 44.6 100% 86.49 100%
49
Table 15: Gas line classes under each suitability types
Proximity to Gasline
Suitability Type Within 250 m Within 500 m Within 750 m
Above 750 m (not Suit.)
Area (%) Area (%) Area (%) Area (%) Highly Suitable 15.3 42% 4.68 15% 3.15 12% 16.11 8% Moderately Suitable 18.72 51% 15.03 49% 11.8 44% 61.2 29% Low Suitable 1.44 4% 8.82 29% 10.8 40% 108.36 51% Not Suitable 1.08 3% 1.89 6% 1.35 5% 26.82 13% Total 21.24 100% 25.74 100% 27.1 100% 212.49 100%
Figure 23: Comparison between suitability types and road criteria
50
Figure 24: Comparison between suitability types and electricity criteria
Figure 25: Comparison between suitability types and gas line criteria
51
6.1.1.4 Community Characteristics
Table 16: Proximity to school under each suitability types
Proximity to School
Suitability Type Within 250 m Within 500 m Within 750 m
Above 750 m (not Suit.)
Area (%) Area (%) Area (%) Area (%) Highly Suitable 12.69 15% 7.2 14% 10.53 15% 8.82 9% Moderately Suitable 32.4 39% 27.18 54% 22.77 32% 24.39 24% Low Suitable 33.66 40% 14.31 28% 31.05 43% 50.4 50% Not Suitable 5.22 6% 2.07 4% 7.38 10% 16.47 16% Total 71.28 100% 43.56 100% 71.73 100% 100.08 100%
According to Table-16, the area coverage of schools at different distances is considerably
higher in moderately suitable and low suitable areas. Actually, schools are distributed all over
the study area to facilitate the education services to all. So, the area coverage of schools is
higher in those two suitable areas, because, those two suitable areas cover major portion
(77%) of the study area. However, within this scope of study, it could not be concluded that
number of schools or their capacities are enough. Rather, it can be said, for example, highly
suitable area has access to schooling facilities, so, residential land-use could be proposed
there. Although the figures in number are somehow different, but similar discussions are also
applicable for the criteria of healthcare services (Table-17).
Table 17: Proximity to healthcare services under each suitability types
Proximity to Healthcare
Suitability Type Within 250 m Within 500 m Within 750 m
Above 750 m (not Suit.)
Area (%) Area (%) Area (%) Area (%) Highly Suitable 2.97 11% 5.94 10% 17.01 13% 13.32 15% Moderately Suitable 15.3 56% 21.51 38% 42.12 32% 27.81 30% Low Suitable 7.92 29% 26.55 46% 58.05 44% 36.9 40% Not Suitable 0.9 3% 3.15 6% 13.68 10% 13.41 15% Total 24.12 100% 51.21 100% 130.9 100% 91.44 100%
52
Figure 26: Comparison between suitability types and school criteria
Figure 27: Comparison between suitability types and healthcare criteria
53
Table 18: Proximity to bus stop under each suitability types
Proximity to Bus Stop
Suitability Type Within 250 m Within 500 m
Insignificant or Not Suit.
Area (%) Area (%) Area (%) Highly Suitable 0.9 29% 4.95 23% 33.39 12% Moderately Suitable 1.62 53% 11.34 53% 93.78 33% Low Suitable 0.45 15% 3.51 16% 125.5 44% Not Suitable 0.09 3% 1.71 8% 29.34 10% Total 2.16 100% 16.56 100% 282 100%
According to Table-18, areas that cover upper than 500 meter distances from bus tops are not
significant. So, these areas have been added with Not Suit. areas. Those areas are
insignificant and were added with Not Suit. as their weightage value is close to zero. Within
250 meter distance, area coverage is highest in moderately suitable area (53%) followed by
highly suitable area (29%) which are also almost similar for areas within 500 meter distance.
So, it can be inferred that highly and moderately suitable areas have greater accessibility to
bus stops that is an advantage for residential, commercial and industrial development.
Figure 28: Comparison between suitability types and bus stop criteria
54
Table 19: Proximity to market/bazar under each suitability types
Proximity to Market/ Bazar
Suitability Type Within 250 m Within 500 m Within 750 m
Above 750 m (not Suit.)
Area (%) Area (%) Area (%) Area (%) Highly Suitable 4.59 21% 5.76 18% 5.85 16% 23.04 11% Moderately Suitable 11.61 54% 13.77 44% 15.21 42% 66.15 30% Low Suitable 4.41 21% 10.35 33% 12.42 35% 102.24 47% Not Suitable 0.9 4% 1.71 5% 2.52 7% 26.01 12% Total 16.92 100% 25.83 100% 36 100% 217.44 100%
As Table-19 shows, some 217 sq. km of the study area is Not Suit. for access to market
places. In fact, these market places are considerably big. People, who stay in long distances
from these market places, don’t come here in daily basis. They come in these market places
one or two times in a week. They buy their daily necessary products from local shops or local
small markets or temporary Bazars. This criterion has value on overall weight of suitability,
but, as an individual criterion nothing could be inferred for land-use planning.
Figure 29: Comparison between suitability types and market/bazar criteria
55
Table 20: Proximity to park under each suitability types
Proximity to Park Suitability Type
Within 250 m Within 500 m Insignificant or Not Suit. Area (%) Area (%) Area (%)
Highly Suitable 0.54 3% 5.94 6% 32.76 17% Moderately Suitable 4.32 20% 30.87 33% 71.55 37% Low Suitable 13.32 62% 40.77 43% 75.33 39% Not Suitable 3.24 15% 16.29 17% 11.61 6% Total 20.88 100% 87.93 100% 191.25 100%
According to Table-20, similar to bus stop, proximity to park got little preferences over other
criteria. So, areas within 500 to 750 distances from park showed as insignificant as their
weight value is close to zero. 62% area within 250 meter and 43% area within 250 to 500
meter fall under the category of low suitable land, that is, most significant areas close to park
lie on low suitable areas. Actually, within the whole study area, no data of park was found.
The study area has no planned development that can be a reason for not having any area
designated for park. As stated in previous chapter, some reserved open spaces have been
treated as park in this study, in a sense, these areas could be used for park in future. The result
in the table validates that intention. Low suitable area, which is not useful for hard urban
development as discussed previously, should be used for soft development, like creating
recreational facilities. Parks along with other recreational facilities could be developed in the
reserved open spaces, which belong to the low suitable area.
Figure 30: Comparison between suitability types and park criteria
56
6.1.2 Land-use Proposal
According to the above discussions, the dominant features under each suitability type can be
shown like as following Table-21 –
Table 21: Dominant features under each suitability types
Suitability Type Dominant Features
Highly Suitable
Vacant Land, High Elevation, No earthquake risk, Close to utility
services, close to bus stops
Moderately Suitable
Homestead, Vacant Land, residential, High Elevation, Minimum
earthquake risk, close to utility services, close to bus stops, close to
market places
Low Suitable Agriculture, homestead, earthquake risk, close to parks
Not Suitable Agriculture, earthquake risk
Considering the above dominant features in Table-21, according to each suitability type we
can prepare the following land-use proposal –
6.1.2.1 Land-use in Highly Suitable Area
The highly suitable areas are featured by vacant land, high elevation, connected with utilities
and community services. Hard urban development can be promoted in these areas. These
areas should be used for urban residential areas to facilitate new housing development.
6.1.2.2 Land-use in Moderately Suitable Area
Moderately suitable areas are featured by homestead, vacant land, existing residential use,
high elevation, minimum earthquake risk, close to market places and connected with utilities
and community services. Within this scope of study, these areas can be proposed as mixed-
use area. This mixed-use area has potentiality of housing, commercial and industrial
development. Detail study is necessary to find out which part of this area is preferable for
industrial development. Other area could be used for both residential and commercial
development as such mixed-use development is appreciated in Bangladesh. Any hard urban
development should be protected in earthquake risk zone.
6.1.2.3 Land-use in Low Suitable Area
Low suitable areas are predominated by agricultural and low lying flood flow zones. These
areas should be free from major development. The lands here are close to fault lines. So, the
places are also prone to earth quack risk. Here development controlling mechanism is earnest
57
necessary. Agricultural lands should be protected as Bangladesh is facing acute shortages of
agricultural land. However, as these areas fall within the Greater Dhaka Metropolitan Area
jurisdiction, so, some parcels of land in these areas could be used for sports grounds, parks
and other recreational open spaces as a requirement of city’s civic amenities.
6.1.2.4 Land-use in Not Suitable Area
The Not Suitable Area is featured by agricultural uses, restricted areas and high earthquake
risk. Any type of use other than agriculture should be protected in this area.
6.1.3 Comparison with DAP
DAP was prepared by a number of experts with their supporting team, used huge technical
and financial resources as well as time. So, it could be deemed as a hypothetical experiment
to compare the DAP with this minor study. However, if we follow the DAP report, around
9.16% of total planning area of Group-E has been delineated for Urban Residential Zone. The
highly suitable area, according to this study, is 13% of Group-E. Within this highly suitable
area, some portion of land would be allocated for road networks and other civic facilities. In
that sense, treating this 13% area as urban residential zone shows a considerable result
comparing to DAP report. Again, it should be mentioned that according to DAP report,
around 52% of Group-E planning area was proposed for reserving as agricultural and flood
flow areas. In this study result, the low suitable area and not suitable areas are 52% of Group-
E area in together. The study suggests using these areas as agricultural use with minor use of
recreational services. Here, the result is similar to DAP’s proposition. At least, this 52% area
could be excluded from DAP preparation processes, thus, could minimize the resources use.
Then, it could be followed more focused development plan to moderately suitable area and
highly suitable area, where minor study could be enough for highly suitable area as it already
have shown the suitability of residential use. So, with these thesis findings, it could be
suggested that GIS-based AHP could be a useful tool for urban land-use planning, where the
tool would assist the decision makers and planners for taking planning decision quickly,
within limited resources and in a consistent and structured way. However, the thesis
suggested the land-use plan at a regional scale that could be utilized successfully for structure
plan or similar strategic planning.
6.2 Conclusion
In this study, a geodatabase has been conceptualized at first for urban land-use planning using
UML model, where five broad themes; like administrative and planning boundary, land-use,
58
physical and geological, utilities, community facilities and demography; were considered.
Such conceptual model has specified data requirements and database structure to prepare
land-use plan for DMDP that will enable the authority to administer the land-use plan.
Then, this study prepared an urban land-use plan using GIS and AHP at a regional scale,
where data used was not in very detail. Such planning approach would be much helpful for
the planners to make structure plan and to adopt other strategic planning. A regional planning
becomes successful when it can be integrated well with the local level or detail level
planning. This research tried to validate its regional urban planning approach with the Detail
Area Plan and found success in some of the cases. This validation pronounces the suitability
of GIS-based AHP application in land-use planning. Such regional approach will select the
areas where more intensive development can be allowed and will identify the areas where
development control should be imposed. In this manner, the specific areas could be chosen
for detail level planning, like DAP of DMDP, using available data. As a result, entire works
would be minimized by reducing detail physical and topographical survey for all over the city
region; more focused development could be promoted; and finally, financial and technical
resource uses could be minimized. However, further research is necessary for getting more
justification in this regard.
59
References
1. Aczel, J. and Satty, T., 1983. Procedures for synthesizing ratio judgments. Journal of
Mathemetical Psychology, 27(1), 93-102. doi: 10.1016/0022-2496(83)90028-7
2. Anderson, J., 1976. A Land use and Land Cover Classification System with Remote
Sensor Data . Government Printing Office, Washington, D.C., U.S.
3. (BBS) Bangladesh Bureau of Statistics, 2001. Community Series, Zila-Dhaka.
4. Booch, G., Rumbaugh, J. and Jacobson, I., 2005. The Unified Modelling Language
User Guide. Addison Wesley.
5. Burley, T., 1961. Land use or land utilization? Professional Geographer, 13(6), 18-
20. doi:10.1111/j.0033-0124.1961.136_18.x
6. Closa, D., Gardiner, A., Giemsa, F. and Machek, J., 2010. Database and database
management system. In: Parent Law for Computer Scienctists. Springer, Berlin, 75-
87.
7. Dai, F., Lee, C. and Zhang, X., 2001. GIS-based geo-environmental evaluation for
urban land-use planning: a case study. Engineering Geology, 61, 257-271. doi:
10.1016/S0013-7952(01)00028-X.
8. Department for Communities and Local Government: London, 2009. Multi-Criteria
Analysis: A Manual . Retrieved February 4, 2012, from
http://eprints.lse.ac.uk/12761/1/Multicriteria-criteria_Analysis.pdf
9. Dong, J., Zhuang, D., Xu, X. and Ying, L., 2008. Integrated evaluation of urban
development suitability based on Remote Sensing and GIS Techniques – A Case
Study in Jingjinji area, China. Sensor, 5975-5986. doi: 10.3390/s8095975
10. Dong, Y., Xu, Y., Li, H. and Dai, M., 2008. Decision support: A comparative study of
the numerical scales and the prioritization methods in AHP. European journal of
operation research, 186, 229-242. doi: 10.1016/j.ejor.2007.01.044
11. Dong, Y., Zhang, G., Hong, W. C. and Xu, Y., 2010. Consensus models for AHP
group decision making under row geometric mean prioritization method. Dicision
Support Systems, 49, 281-289. doi: 10.1016/j.dss.2010.03.003
12. Ernest, F. and Kirti, P., 1998. Aggregating individual judgments and priorities with
the Analytical Hierarchy Process. European Journal of Operational Research, 108,
165-169. doi: 10.1016/S0377-2217(97)00244-0
60
13. FAO (Food and Agricultural Organization), 1996. Guidelines for land-use planning
FAO, UNO. Rome.
14. GOB (Government Of Bangladesh), Ministry of Housing and Public Works, Planning
Section-3, Gazette, 22 June 2010.
15. Harker, P. and Vargas, L., 1987. The theory of ratio scale estimation: Saaty's
Analytical Hierarchy Process. Management Science, 33(11), 1383-1403. doi:
org/10.1287/mnsc.33.11.1383
16. Islam, I., Sharmin, M., Masud, F. and Moniruzzaman, M., 2009. Urban consolidation
approach for Dhaka city: prospects and contraints. Global Built Environment Review,
7 (1), 50-68.
17. Jacobson, I., Ericson, M. and Jacobson, A., 1994. The Object Advantage Business
Process Reengineering with Object Technology. Addison Wesley.
18. Jeer, S., 2001. Land-based Classification Standards . American Planning Association,
Chicago.
19. Kahraman, C., 2008. Multi-Criteria decision making methods and fuzzy sets. In:
Kahraman, C., (ed.), Fuzzy Multi-Criteria Decision Making. Springer Science +
Business Media, 1.
20. Lisbao Filho, J. and Iochpe, C., 1999. Specifying analysis patterns for geographic
databases on the basis of conceptual framework. In: ACM Symposium on Advances in
Geographic Information Systems , Kansas. 9-13.
21. Lisbao Filho, J. and Iochpe, C., 2008. Modeling with a UML profile. In: Shekhar, S.
and Xiong, H (eds.), Encyclopedia of GIS. Verlag: Springer. 691-700.
22. Lootsma, F., 1993. Scale sensitivity in multiple AHP and SMART. Journal of Multi-
Criteria Decision Analysis, 2, 87-110. doi: 10.1002/mcda.4020020205
23. Lovelace, E., 1993. Harland Bartholomew: His Contributions to American Urban
Planning . University of Illions Office of Printing Services, Urbana, IL.
24. Malczewski, J., 2004. GIS-based land-use suitability analysis: a critical overview.
Progress in planning, 62, 3-65. doi: 10.1016/j.progress.2003.09.002
25. Nallathiga, R., 2009. From master plan to vision plan: the changing role of plans and
plan making in the city development (with reference to Mumbai). TERUM, 4(13),
141-157.
26. OMG. (n.d.). Retrieved January 21, 2013, from http://www.uml.org
27. RAJUK, Ministry of Housing and Public Work, Government of the People’s Republic
of Bangladesh, 2010. Preparation of Detailed Area Plan for Group-E of DMDP.
61
Retrieved November 26, 2012, from
http://www.rajukdhaka.gov.bd/rajuk/image/dap/groupE_Report/Chapters_e.pdf
28. Ramanathan, R. and Ganesh, L., 1994. Group preference aggregation methods
employed in AHP: An evaluation and an intrinsic process for deriving members'
weightages. European Journal of Operational Research, 79, 149-265. doi:
10.1016/0377-2217(94)90356-5
29. Rumbaugh, J., Jacobson, I. and Booch, G., 1999. The Unified Modelling Language
Reference Manual . Addison Wesley .
30. Saaty, T., 1980. The Analytical Hierarchy Process. McGraw-Hill International, New
York .
31. Saaty, T., 1990. How to make a decision: the analytical hierarchy process. European
journal of operation research, 48, 9-26. doi: 10.1016/0377-2217(90)90057-I
32. Saaty, T. and Vargas, L., 2001. Model, methods, concepts and application of
analytical hirarchy process. In: Saaty, T. and Vargas, L. (eds.), International Series in
Operations, Research and Management Sciences. Kluwer Academic Publishers,
Boston/Dodrecht/London.
33. Shekhar, S. and Chawla, S., 2003. Spatial Database . Prentice Hall, New York.
34. Tudes, S. and Yigiter, N.D., 2010. Preparation of land use planning model using GIS
based on AHP: case study Adana-Turkey. Bulletin of engineering geology and the
environment . 69, 235-245. doi: 10.1007/s10064-009-0247-5
35. Ullah, K. M. and Hafiz, R., 2014. Finding suitable locations for ecotourism
development in Cox's Bazar using Geographical Information System and Analytical
Hierarchy Process. Geocarto International. 29(3), 256-267 doi:
10.1080/10106049.2012.760005
36. United States Department of the Interior Bureau of Land Management, 2005. Land
Use Planning Handbook.
37. Urban Renewal Administration; Housing and Home Finance Agency; Bureau of
Public Roads; Department of Commerce, 1965. Standard Land Use Coding Manual:
A Standard System of Identifying and Coding Land Use Activities. Government
Printing Office, Washington, DC.
38. Van Lier, H., 1998. The role of land use planning in sustainable rural systems.
Landscape and Urban Planning, 41(2), 83-91. doi: 10.1016/S0169-2046(97)00061-3
39. Wang, X. and Hofe, R. V., 2007. Research Methods in Urban and Regional Planning
. Springer, Berlin Heidelberg/New York.
62
40. World Bank, 2007. Dhaka: Improving Living COndition for the Urban Poor:
Bangladesh Development Series. The World Bank Office, Dhaka.
41. Yeung, A. K. and Hall, G. B., 2007. Spatial Database Systems, 87. Springer,
Dordrecht.
42. Zhang, Y., Li, A. and Fung, T., 2012. Using GIS and Multi-criteria decision analysis
for conflict resolution in land use planning. Procedia Environmental Sciences, 13,
2264-2273. doi: 10.1016/j.proenv.2012.01.215
63
Appendix 1: Comparison Matrixes
Factor Matrixes
Expert 1:
Land-
use Physical/Geological Utilities Community Facilities
Land-use 1.00 2.00 7.00 8.00
Physical/geological 0.50 1.00 6.00 7.00
Utilities 0.14 0.17 1.00 2.00
Community Facilities 0.13 0.14 0.50 1.00
Expert 2:
Land-
use Physical/Geological Utilities Community Facilities
Land-use 1.00 2.00 5.00 7.00
Physical/geological 0.50 1.00 5.00 7.00
Utilities 0.20 0.20 1.00 2.00
Community
Facilities 0.14 0.14 0.50 1.00
Expert 3:
Land-
use Physical/Geological Utilities Community Facilities
Land-use 1.00 1.00 4.00 8.00
Physical/geological 1.00 1.00 4.00 8.00
Utilities 0.25 0.25 1.00 3.00
Community
Facilities 0.13 0.13 0.33 1.00
64
Expert 4:
Land-
use Physical/Geological Utilities Community Facilities
Land-use 1.00 2.00 4.00 8.00
Physical/geological 0.50 1.00 3.00 8.00
Utilities 0.25 0.33 1.00 3.00
Community Facilities 0.13 0.13 0.33 1.00
Expert 5:
Land-
use Physical/Geological Utilities Community Facilities
Land-use 1.00 3.00 5.00 8.00
Physical/geological 0.33 1.00 3.00 7.00
Utilities 0.20 0.33 1.00 3.00
Community Facilities 0.13 0.14 0.33 1.00
Criteria Matrixes
Land-use
Expert 1:
Agriculture Homestead Residential Vacant Land
Agriculture 1.00 0.25 2.00 0.33
Homestead 4.00 1.00 4.00 3.00
Residential 0.50 0.25 1.00 0.20
Vacant land 3.00 0.33 5.00 1.00
Expert 2:
Agriculture Homestead Residential Vacant Land
Agriculture 1.00 0.20 2.00 0.17
Homestead 5.00 1.00 4.00 0.50
Residential 0.50 0.25 1.00 0.20
Vacant land 6.00 2.00 5.00 1.00
65
Expert 3:
Agriculture Homestead Residential Vacant Land
Agriculture 1.00 0.17 2.00 0.14
Homestead 6.00 1.00 4.00 0.50
Residential 0.50 0.25 1.00 0.17
Vacant land 7.00 2.00 6.00 1.00
Expert 4:
Agriculture Homestead Residential Vacant Land
Agriculture 1.00 0.20 3.00 0.20
Homestead 5.00 1.00 5.00 1.00
Residential 0.33 0.20 1.00 0.13
Vacant land 5.00 1.00 8.00 1.00
Expert 5:
Agriculture Homestead Residential Vacant Land
Agriculture 1.00 0.17 2.00 0.20
Homestead 6.00 1.00 5.00 2.00
Residential 0.50 0.20 1.00 0.13
Vacant land 5.00 0.50 8.00 1.00
Utilities
Expert 1:
Road Gas Supply Electricity
Road 1.00 8.00 3.00
Gas Supply 0.13 1.00 0.25
Electricity 0.33 4.00 1.00
Expert 2:
Road Gas Supply Electricity
Road 1.00 9.00 2.00
Gas Supply 0.11 1.00 0.25
Electricity 0.50 4.00 1.00
66
Expert 3:
Road Gas Supply Electricity
Road 1.00 7.00 2.00
Gas Supply 0.14 1.00 0.20
Electricity 0.50 5.00 1.00
Expert 4:
Road Gas Supply Electricity
Road 1.00 7.00 2.00
Gas Supply 0.14 1.00 0.20
Electricity 0.50 5.00 1.00
Expert 5:
Road Gas Supply Electricity
Road 1.00 6.00 2.00
Gas Supply 0.17 1.00 0.20
Electricity 0.50 5.00 1.00
Physical-Geological
Expert 1:
Elevation Fault line
Elevation 1.00 2.00
Fault line 0.50 1.00
Expert 2:
Elevation Fault line
Elevation 1.00 0.50
Fault line 2.00 1.00
Expert 3:
Elevation Fault line
Elevation 1.00 0.50
Fault line 2.00 1.00
67
Expert 4:
Elevation Fault line
Elevation 1.00 0.50
Fault line 2.00 1.00
Expert 5:
Elevation Fault line
Elevation 1.00 0.50
Fault line 2.00 1.00
Community Facilities
Expert 1:
School Healthcare Centre Bust Stop Market/Bazar Park
School 1.00 5.00 7.00 4.00 8.00
Healthcare Centre 0.20 1.00 3.00 0.50 5.00
Bus Stop 0.14 0.33 1.00 0.25 3.00
Market/Bazar 0.25 2.00 4.00 1.00 5.00
Park 0.13 0.20 0.33 0.20 1.00
Expert 2:
School Healthcare Centre Bust Stop Market/Bazar Park
School 1.00 2.00 5.00 4.00 6.00
Healthcare Centre 0.50 1.00 3.00 3.00 4.00
Bus Stop 0.20 0.33 1.00 1.00 2.00
Market/Bazar 0.25 0.33 1.00 1.00 2.00
Park 0.17 0.25 0.50 0.50 1.00
Expert 3:
School Healthcare Centre Bust Stop Market/Bazar Park
School 1.00 3.00 5.00 6.00 8.00
Healthcare Centre 0.33 1.00 3.00 4.00 6.00
Bus Stop 0.20 0.33 1.00 2.00 2.00
Market/Bazar 0.17 0.25 0.50 1.00 1.00
Park 0.13 0.17 0.50 1.00 1.00
68
Expert 4:
School Healthcare Centre Bust Stop Market/Bazar Park
School 1.00 3.00 6.00 5.00 7.00
Healthcare Centre 0.33 1.00 3.00 2.00 6.00
Bus Stop 0.17 0.33 1.00 0.50 2.00
Market/Bazar 0.20 0.50 2.00 1.00 3.00
Park 0.14 0.17 0.50 0.33 1.00
Expert 5:
School Healthcare Centre Bust Stop Market/Bazar Park
School 1.00 4.00 6.00 5.00 7.00
Healthcare Centre 0.25 1.00 2.00 2.00 6.00
Bus Stop 0.17 0.50 1.00 0.50 2.00
Market/Bazar 0.20 0.50 2.00 1.00 3.00
Park 0.14 0.17 0.50 0.33 1.00
69
Appendix 2: Factor and Criteria Weights
Factor Weight
Factor Expert1 Expert2 Expert3 Expert4 Expert5
Geometric
Mean
Normalized
Value
Land-use 0.52 0.50 0.42 0.50 0.56 0.50 0.50
Physical/geological 0.35 0.35 0.42 0.33 0.28 0.34 0.34
Utilities 0.08 0.09 0.12 0.12 0.12 0.10 0.10
Community
Facilities 0.05 0.06 0.05 0.05 0.05 0.05 0.05
Sum 0.99 1.00
Criteria Weight
Land-use
Factor Expert1 Expert2 Expert3 Expert4 Expert5
Geometric
Mean
Normalized
Value
Agriculture 0.12 0.10 0.09 0.09 0.09 0.10 0.10
Homestead 0.50 0.32 0.32 0.32 0.48 0.38 0.39
Residential 0.08 0.08 0.07 0.07 0.06 0.07 0.07
Vacant land 0.29 0.50 0.52 0.52 0.37 0.43 0.44
Sum 0.98 1.00
Utilities
Factor Exeprt1 Exeprt2 Exeprt3 Exeprt4 Exeprt5
Geometric
Mean
Normalized
Value
Road 0.67 0.67 0.67 0.67 0.67 0.67 0.67
Gas
Supply 0.07 0.07 0.07 0.07 0.07 0.07 0.07
Electricity 0.26 0.26 0.26 0.26 0.26 0.26 0.26
Sum 1.00 1.00
Physical-Geological
Factor Expert1 Expert2 Expert3 Expert4 Expert5
Geometric
Mean
Normalized
Value
Elevation 0.67 0.33 0.33 0.33 0.33 0.38 0.40
Fault line 0.33 0.67 0.67 0.67 0.67 0.58 0.60
Sum 0.96 1.00
70
Community Facilities
Factor Member1 Member2 Member3 Member4 Member5
Geometric
Mean
Normalized
Value
School 0.53 0.46 0.51 0.51 0.53 0.51 0.52
Healthcare
Centre 0.15 0.27 0.26 0.23 0.20 0.22 0.22
Bus Stop 0.07 0.10 0.11 0.08 0.08 0.09 0.09
Market/Bazar 0.21 0.11 0.06 0.13 0.13 0.12 0.12
Park 0.04 0.06 0.06 0.05 0.05 0.05 0.05
Sum 0.98 1.00
71
Series from Lund University
Department of Physical Geography and Ecosystem Science
Master Thesis in Geographical Information Science (LUMA-GIS)
1. Anthony Lawther: The application of GIS-based binary logistic regression for slope failure susceptibility mapping in the Western Grampian Mountains, Scotland. (2008).
2. Rickard Hansen: Daily mobility in Grenoble Metropolitan Region, France. Applied GIS methods in time geographical research. (2008).
3. Emil Bayramov: Environmental monitoring of bio-restoration activities using GIS and Remote Sensing. (2009).
4. Rafael Villarreal Pacheco: Applications of Geographic Information Systems as an analytical and visualization tool for mass real estate valuation: a case study of Fontibon District, Bogota, Columbia. (2009).
5. Siri Oestreich Waage: a case study of route solving for oversized transport: The use of GIS functionalities in transport of transformers, as part of maintaining a reliable power infrastructure (2010).
6. Edgar Pimiento: Shallow landslide susceptibility – Modelling and validation (2010).
7. Martina Schäfer: Near real-time mapping of floodwater mosquito breeding sites using aerial photographs (2010)
8. August Pieter van Waarden-Nagel: Land use evaluation to assess the outcome of the programme of rehabilitation measures for the river Rhine in the Netherlands (2010)
9. Samira Muhammad: Development and implementation of air quality data mart for Ontario, Canada: A case study of air quality in Ontario using OLAP tool. (2010)
10. Fredros Oketch Okumu: Using remotely sensed data to explore spatial and temporal relationships between photosynthetic productivity of vegetation and malaria transmission intensities in selected parts of Africa (2011)
11. Svajunas Plunge: Advanced decision support methods for solving diffuse water pollution problems (2011)
72
12. Jonathan Higgins: Monitoring urban growth in greater Lagos: A case study using GIS to monitor the urban growth of Lagos 1990 - 2008 and produce future growth prospects for the city (2011).
13. Mårten Karlberg: Mobile Map Client API: Design and Implementation for Android (2011).
14. Jeanette McBride: Mapping Chicago area urban tree canopy using color infrared imagery (2011)
15. Andrew Farina: Exploring the relationship between land surface temperature and vegetation abundance for urban heat island mitigation in Seville, Spain (2011)
16. David Kanyari: Nairobi City Journey Planner An online and a Mobile Application (2011)
17 Laura V. Drews: Multi-criteria GIS analysis for siting of small wind power plants - A case study from Berlin (2012)
18 Qaisar Nadeem: Best living neighborhood in the city - A GIS based multi criteria evaluation of ArRiyadh City (2012)
19 Ahmed Mohamed El Saeid Mustafa: Development of a photo voltaic building rooftop integration analysis tool for GIS for Dokki District, Cairo, Egypt (2012)
20 Daniel Patrick Taylor: Eastern Oyster Aquaculture: Estuarine Remediation via Site Suitability and Spatially Explicit Carrying Capacity Modeling in Virginia’s Chesapeake Bay (2013)
21 Angeleta Oveta Wilson: A Participatory GIS approach to unearthing Manchester’s Cultural Heritage ‘gold mine’ (2013)
22 Ola Svensson: Visibility and Tholos Tombs in the Messenian Landscape: A Comparative Case Study of the Pylian Hinterlands and the Soulima Valley (2013)
23 Monika Ogden: Land use impact on water quality in two river systems in South Africa (2013)
24 Stefan Rova: A GIS based approach assessing phosphorus load impact on Lake Flaten in Salem, Sweden (2013)
25 Yann Buhot: Analysis of the history of landscape changes over a period of 200 years. How can we predict past landscape pattern scenario and the impact on habitat diversity? (2013)
73
26 Christina Fotiou: Evaluating habitat suitability and spectral heterogeneity models to predict weed species presence (2014)
27 Inese Linuza: Accuracy Assessment in Glacier Change Analysis (2014)
28 Agnieszka Griffin: Domestic energy consumption and social living standards: a GIS analysis within the Greater London Authority area (2014)
29 Brynja Guðmundsdóttir Detection of potential arable land with remote sensing and GIS - A Case Study for Kjósarhreppur (2014)
30 Oleksandr Nekrasov Processing of MODIS Vegetation Indices for analysis of agricultural droughts in the southern Ukraine between the years 2000-2012 (2014)
31 Sarah Tressel Recommendations for a polar Earth science portal
in the context of Arctic Spatial Data Infrastructure (2014)
32 Caroline Gevaert Combining Hyperspectral UAV and Multispectral Formosat-2 Imagery for Precision Agriculture Applications (2014).
33 Salem Jamal-Uddeen Using GeoTools to implement the multi-criteria evaluation analysis - weighted linear combination model (2014)
34 Samanah Seyedi-Shandiz Schematic representation of geographical railway network at the Swedish Transport Administration (2014)
35 Kazi Masel Ullah Urban Land-use planning using Geographical Information System and analytical hierarchy process: case study Dhaka City (2014)
